Novel Effect of Oxidized Low-Density Lipoprotein
http://www.100md.com
Sergiy Sukhanov, Yusuke Higashi, Shaw-Yu
参见附件。
the Cardiology Section (S.S., Y.H., S.-Y.S, P.D.), Department of Medicine, Tulane University, New Orleans, La
Department of Biological Chemistry (H.I.), Showa University, Tokyo, Japan
Innoshima City Medical Associated Hospital (K.O.), Japan
Pathology Department (S.P.), Louisiana State University, New Orleans.
Abstract
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is a classical glycolytic enzyme that is involved in cellular energy production and has important housekeeping functions. We used the natural prooxidant and proatherogenic molecule oxidized low-density lipoprotein (OxLDL) to determine a potential link between OxLDL-promoted oxidative stress, GAPDH expression, and smooth muscle cell energy metabolism. OxLDL but not native LDL (nLDL) produced a 60% to 100% dose- and time-dependent reduction of GAPDH protein. OxLDL increased reactive oxygen species (ROS) formation, including rapid elevation of H2O2 levels. OxLDL decreased intracellular catalase expression, likely contributing to the increase in H2O2. Antioxidants, anti-CD36 receptor antibody, NADPH oxidase, or lipoxygenase blockers decreased OxLDL-specific ROS and prevented GAPDH downregulation. 12/15-Lipoxygenase or p47phox deficiency resulted in attenuation of GAPDH downregulation, but 5-lipoxygenase suppression had no effect. OxLDL or exogenous H2O2 oxidized GAPDH thiols, decreasing GAPDH protein half-life and increasing GAPDH sensitivity to proteasome-mediated protein degradation in vitro. OxLDL- or small interfering RNA–specific downregulation of GAPDH resulted in 65% reduction in glycolysis rate and 82% decrease in ATP levels. In conclusion, our data demonstrate that OxLDL downregulated GAPDH via a H2O2-dependent decrease in protein stability. GAPDH protein damage resulted in marked depletion of cellular ATP levels. Our data have important implications for understanding the metabolic effect of OxLDL on the vessel wall and mechanism of atherogenesis.
Key Words: reactive oxygen species low-density lipoprotein energy metabolism vascular smooth muscle
Introduction
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (EC 1.2.1.12) is a glycolytic enzyme involved in energy production and is one of the most abundant cellular proteins. GAPDH is constitutively expressed in different cells and is viewed as an important housekeeping molecule. However, it has become clear that the role of GAPDH is not restricted to its pivotal glycolytic function. Indeed, GAPDH regulates multiple other processes, including membrane fusion, nuclear RNA transport, and DNA replication and repair.1
Increasing evidence suggests that GAPDH is a preferential target for oxidative stress. Nitric oxide (NO) and reactive oxygen species (ROS)-mediated oxidation of GAPDH thiols (S-thiolation) and subsequent inhibition of enzyme activity have been described in colon epithelial cells during inflammatory bowel disease.2 Wentzel et al3 demonstrated that diabetic embryopathy in the rat is associated with the inhibition of GAPDH activity and a decrease in GAPDH mRNA and that this effect is blunted by the antioxidant N-acetylcysteine. Taken together, these observations indicate a high sensitivity of GAPDH to cell redox status and potential involvement of this enzyme in oxidative stress-promoted events.
Oxidized low-density lipoprotein (OxLDL) is a powerful natural prooxidant derived in vivo from nLDL by cell-mediated oxidation.4 OxLDL is a key proatherogenic agent5 that induces, as we have shown, an increase in ROS formation, specific alterations in gene/protein expression, and apoptosis of vascular smooth muscle cells.6–9 We designed this study to determine a potential link between OxLDL-specific increase in ROS, GAPDH expression, and human aortic smooth muscle cell (HASMC) energy metabolism. Our findings indicate dramatic OxLDL-induced downregulation of GAPDH via a redox-dependent signal pathway, resulting in depletion of cellular energy stores. These results provide a potential mechanism for understanding metabolic alterations of the vasculature during the process of atherogenesis.
Materials and Methods
Cell culture, isolation, oxidative modification of LDL, and immunoblotting analysis protocols have been described previously.6 Formation of intracellular ROS were assessed with 6-carboxy-2',7'-dichlorodihydro-fluorescein diacetate (CDC-H2F), hydroethidine, and Amplex Red kit (all from Molecular Probes). GAPDH free thiol status was measured by labeling of cell extracts with the thiol-sensitive fluorescent dye fluorescein-5-maleimide (Molecular Probes), and metabolic labeling of HASMC was performed with EXPRE35S35S Protein Labeling Mix (Perkin Elmer) followed by immunoprecipitation with anti-GAPDH (Chemicon) or anti–-actin (Sigma) antibody. An expanded Materials and Methods section containing details regarding ROS measurements, GAPDH purification, proteasome protein-degradation assay, and use of small interfering RNA (siRNA), as well as Figures I through III, are given in the data supplement, available at http://circres.ahajournals.org.
Results
OxLDL Downregulates GAPDH Protein
Confluent (80% to 90%) HASMC were treated with 60 μg/mL OxLDL or native LDL (nLDL) (control) for 0 to 24 hours or incubated in serum-free media (SFM). No change in GAPDH expression was found after cell treatment with nLDL or SFM; in contrast, cell incubation with OxLDL time-dependently downregulated GAPDH protein (Figure 1A). GAPDH protein level was reduced in OxLDL-treated cells to 68±15% and 33±10% compared with control at 9 and 15 hours, respectively, and GAPDH expression at 24 hours was virtually not detectable.
HASMC treatment by LDLs for 15 hours was selected for generation of dose-response curves and to assess effects of antioxidants. For dose-response experiments, HASMC were treated with 0 to 100 μg/mL OxLDL or nLDL and GAPDH expression was analyzed by immunoblotting. Sixty micrograms per milliliter OxLDL decreased GAPDH levels to 38±11%, and 80 μg/mL OxLDL downregulated GAPDH to 25±3%. OxLDL (20 to 100 μg/mL) produced no changes in protein levels of other housekeeping proteins -actin and -tubulin (Figure 1B). However, we noted that treatment of HASMC with highly OxLDL (>50 nmol malondialdehyde per milligram protein) slightly decreased -actin levels (15±5% compared with nLDL-treated cells), as evidenced by immunoblotting, together with marked downregulation of GAPDH protein levels.
Immunoblotting data obtained with mouse monoclonal anti-GAPDH antibody were further confirmed with rabbit polyclonal anti-GAPDH antibody (data not shown). We also examined the efficiency of the recognition of potentially oxidized antigen (GAPDH) by cognitive antibody (mouse anti-GAPDH) in a routine Western analysis. Purified chicken muscle GAPDH (0.1 mg/mL; Sigma) was directly oxidized by 1.5 mmol/L H2O2, and an aliquot of oxidized and untreated GAPDH (control) were analyzed by immunoblotting. No differences were found between band density of oxidized GAPDH and control (data not shown). Although the possibility remains that there is a loss of immunoreactivity independent of reduced expression of the GAPDH protein. This appears unlikely in view of our results comparing immunoreactivity of oxidized and nonoxidized GAPDH using 2 different epitope-specific anti-GAPDH antibodies.
GAPDH gene expression was analyzed by Northern blotting. In time-course experiments, GAPDH protein was decreased after 9 hours of OxLDL treatment, and for this period (0 to 9 hours), GAPDH mRNA levels were not changed significantly compared with nLDL (Figure 1C). However, longer incubation with OxLDL (15 to 24 hours) produced a slight decrease of GAPDH mRNA (18±5% at 24 hours).
Antioxidants Suppress OxLDL-Specific Peroxides, Decrease Superoxides, and Prevent Downregulation of GAPDH
Because OxLDL signaling can involve ROS,10 we assessed formation of intracellular peroxides and superoxides in LDL-treated HASMC with the peroxide-sensitive dye CDC-H2F and the superoxide-specific dye hydroethidine (HE). Both oxidized and native LDL (60 μg/mL) increased cell peroxides time dependently up to 9 hours of treatment. For 2 to 6 hours of incubation, OxLDL produced 40% to 70% more peroxides than nLDL, but longer treatments resulted in no difference between peroxide levels induced by OxLDL and control (see supplemental Figure IA). Unlike peroxide formation, superoxide level was significantly higher in cells treated with OxLDL versus nLDL for all time points up to 30 hours. OxLDL increased peroxides dose dependently (20 to 80 μg/mL), with a peak formation at 3 to 4 hours. OxLDL-specific formation of superoxides was also dose-dependent with a maximum at 20 hours (supplemental Figure IB and ID). These time points were chosen for assaying OxLDL-specific peroxide and superoxide generation in further experiments.
Pretreatment with the water-soluble vitamin E derivative Trolox (0.25 mmol/L) or antioxidant Tiron (1 mmol/L) dramatically reduced basal and nLDL-induced peroxides, with complete elimination of OxLDL-specific peroxides (Figure 2A). In contrast, antioxidants were less efficient in inhibition of cell superoxides induced by OxLDL; both nLDL- or OxLDL-induced superoxides were equally suppressed by antioxidants (Figure 2B). Trolox or Tiron completely prevented OxLDL-specific decrease in GAPDH (Figure 2C). Based on this data, we concluded that ROS mediate the OxLDL-specific GAPDH decrease.
CD36 Scavenger Receptor, 12/15-Lipoxygenase, and NADPH Oxidase Mediate OxLDL-Induced ROS Increase and GAPDH Downregulation
The CD36 scavenger receptor mediates the uptake of OxLDL, allowing OxLDL-specific intracellular signaling and development of atherosclerosis.11 Using real time PCR we confirmed the presence of CD36 mRNA in HASMC and one unique protein band immunogenic to CD36 antibody (MW=90±7 kDa, n=3) was detected in cell lysates by immunoblotting (data not shown). To verify the potential involvement of CD36 receptor in OxLDL-specific ROS increase and GAPDH downregulation we used mouse FA6–152 anti-CD36 blocking antibody (Immunotech, Cedex, France), a specific inhibitor of CD36-dependent OxLDL uptake.11 We found that anti-CD36 antibody (20 μg/mL) decreased OxLDL-specific increase in peroxide and superoxide formation, whereas pretreatment with nonspecific antibody (20 μg/mL normal mouse IgG) had no effect (see supplemental Figure II). Antibody-specific inhibition of CD36 scavenger receptor resulted in complete inhibition of OxLDL-induced GAPDH decrease (Figure 3A).
We used allopurinol, the xanthine oxidase inhibitor; NS-398, the specific blocker of cyclooxygenase-2; indomethacin, the nonspecific blocker of cyclooxygenase-1 and cyclooxygenase-2; and rotenone, the inhibitor of mitochondrial complex I, to study potential involvement of these ROS producing systems in OxLDL-dependent increase in HASMC peroxides and superoxides. Preincubation of HASMC with 100 μmol/L allopurinol, 10 μmol/L NS-398, or 10 μmol/L indomethacin did not block OxLDL-specific peroxide or superoxide formation or downregulation of GAPDH, and rotenone was toxic at concentrations of 10 and 100 μmol/L (data not shown).
Lipoxygenases (LOXs), such as 5-LOX, which is a key enzyme in leukotriene production, and 12/15-LOX are a family of lipid peroxidizing enzymes that are another source of ROS in the vasculature.12 We verified the presence of 5- and 12/15-LOX proteins and corresponding enzyme activities in HASMC (data not shown). We found that the LOX inhibitor AA-861 (5 μmol/L) markedly reduced OxLDL-induced ROS production (supplemental Figure II). The same dose of AA-861 completely blocked OxLDL downregulation of GAPDH (Figure 3B). To identify the LOX form that mediates OxLDL-induced ROS production and GAPDH downregulation, we used 5-LOX–specific and nonspecific (control) siRNA-treated HASMC and vascular smooth muscle cells (SMC) obtained from 12/15-LOX–deficient mice13 and wild-type mice (control). All methods involving animals were approved by the Institutional Animal Care and Use Committee (IACUC) of Tulane University Health Sciences Center. siRNA-mediated 5-LOX protein decrease was 65% as measured by immunoblotting analysis (Figure 3D), and there was a similar decrease in 5-LOX mRNA levels as measured by real-time PCR (data not shown). We found that 12/15-LOX deficiency resulted in a 62±7% reduction of OxLDL-induced peroxides; however, 5-LOX suppression had no effect (supplemental Figure III). OxLDL-induced downregulation of GAPDH was significantly blunted in 12/15-LOX–/– vascular SMC (54±14% inhibition compared with wild-type), whereas 5-LOX suppression had no blocking effect on OxLDL-induced GAPDH downregulation (Figure 3E).
The membrane-bound NADPH oxidase is a predominant source of ROS in multiple pathophysiological conditions in vascular smooth muscle.14 We found that apocynin (50 μmol/L), a potent and selective inhibitor of NADPH oxidase, and the structurally unrelated NADPH oxidase blocker AEBSF (1 mmol/L) were less efficient in suppression of OxLDL-specific ROS compared with AA-861 (supplemental Figure II). NADPH oxidase blockers only partially suppressed downregulation of GAPDH (Figure 3C). p47phox is a critical subunit of the NADPH oxidase, and a null mutation of this gene resulted in attenuation of superoxide production in SMC and suppression of atherosclerosis.15 To verify our data obtained with NADPH oxidase inhibitors, we measured peroxides and GAPDH protein levels in vascular SMC from p47phox–/– or wild-type (control) mice exposed to nLDL or OxLDL. We found that p47phox–/– cells produce 82% less peroxides in response to OxLDL compared with wild type (supplemental Figure III). p47phox deficiency completely blocked OxLDL-induced GAPDH downregulation (Figure 3F).
These data suggest involvement of the CD36 receptor, 12/15-LOX, and NADPH oxidase in ROS formation and OxLDL-specific GAPDH downregulation.
H2O2 Scavengers Completely Block OxLDL-Specific GAPDH Downregulation, and Exogenous H2O2 Downregulates GAPDH
H2O2 is a major oxidative component of OxLDL-induced ROS16 and, therefore, could be involved in ROS-mediated GAPDH downregulation. We measured specific H2O2 release from LDL-exposed cells with/without pretreatment with Trolox (0.25 mmol/L), specific H2O2 scavengers catalase (10 kU/mL), or polyethylene glycol–coupled catalase (PEG-catalase) (1 kU/mL) using the Amplex Red assay. OxLDL (60 μg/mL, 3 hours) significantly increased H2O2 levels compared with nLDL or SFM (Figure 4A). Pretreatment with antioxidant or catalase efficiently suppressed OxLDL-induced H2O2 formation to below basal levels. After calibration of the Amplex Red signal with standards (H2O2, 0 to 1000 μmol/L), we estimated that the OxLDL-specific increase in cellular H2O2 level corresponds to 350 to 450 μmol/L H2O2 (data not shown). This is consistent with reports showing that oxidative stress enhanced cellular H2O2 concentration in vivo to the submillimolar range.17 Therefore, we examined GAPDH expression in response to similar doses of exogenous H2O2. We found that HASMC incubation with H2O2 (225 to 450 μmol/L, 16 hours) resulted in a downregulation of GAPDH and that this effect was dose dependent (Figure 4B).
Based on these data, we hypothesized that OxLDL-specific increase in H2O2 is critical for subsequent GAPDH protein suppression and verified this hypothesis using H2O2 scavengers (catalase and PEG-catalase) and superoxide scavengers (superoxide dismutase [SOD] and PEG-SOD). Only catalases completely blocked OxLDL-induced downregulation of GAPDH (Figure 4C and 4D). The blocking effect of SOD was partial: GAPDH levels still decreased in OxLDL-treated cells compared with nLDL (67±7% of control, n=3, P<0.05) after preincubation with this superoxide scavenger. PEG-SOD at 50 U/mL even potentiated GAPDH downregulation and was toxic at higher concentrations. This data suggested that OxLDL-specific increase in H2O2 is responsible for GAPDH downregulation.
OxLDL Rapidly Decreases Catalase Activity in HASMC and Does Not Alter SOD Activity
Two major antioxidant enzyme systems protect cells from oxidative stress by maintaining safe levels of ROS: SODs, enzymes that catalyze conversion of superoxides to H2O2 including intracellular (Mn-SOD, Cu/Zn-SOD) and extracellular (EC-SOD) forms and catalase.18 To clarify the potential mechanism of OxLDL-induced H2O2 elevation, we measured the activity and protein levels of SODs and catalase in HASMC treated by OxLDL.
Immunoblotting analysis showed that the relative amount of catalase was decreased by 58±5% by OxLDL treatment compared with nLDL, but Mn-SOD protein level was not changed (Figure 5A and 5B). Similarly, the functional activity of catalase was significantly reduced (Figure 5C). However, activity of intracellular SOD, as assessed by Cayman Chemical assay and in situ SOD activity detection, was not different between OxLDL- and nLDL-treated cells (data not shown). Additionally, Western blotting revealed that OxLDL did not alter expression of Cu/Zn-SOD (data not shown). We detected activity of an EC-SOD–like protein in conditioned media from LDL-treated cells using in situ assay and found that this activity also was unregulated by OxLDL (data not shown). These data suggest that the reduction in catalase activity and the maintenance of SOD activity could contribute to the increase in intracellular H2O2 in response to OxLDL.
OxLDL or H2O2 Oxidize GAPDH Thiols and Decrease Protein Stability
The GAPDH molecule has 3 easily oxidized SH groups and 2 of them (Cys149 and Cys153) are essential for GAPDH catalytic activity.19 We measured GAPDH thiol status following cell exposure to OxLDL or H2O2. GAPDH-labeling efficiency with the thiol-sensitive dye fluorescein-5-maleimide was dramatically decreased following treatment with OxLDL (60 μg/mL, 3 hours) or, similarly, after incubation with H2O2 (400 μmol/L, 3 hours) without alteration of GAPDH protein levels (Figure 6).
In theory, GAPDH thiol (R-S-H) modification by an oxidant can include: (1) formation of inter- or intramolecular disulphide bands (R-S-S-R); (2) formation of mixed disulphides with a small intracellular thiol, glutathione (R-S-S-R2); or (3) direct oxidation of the Cys thiol groups to sulfenic acid (R-S-O-H). To address the first possibility, we used nonreducing gel electrophoresis to examine whether formation of an intramolecular disulphide bond or a bond between 2 molecules of GAPDH resulted in altered protein migration by nonreducing SDS-PAGE. Neither H2O2 nor OxLDL treatment of cells resulted in altered GAPDH electrophoretic mobility (data not shown). To address the second possibility, we examined GAPDH thiol oxidation by H2O2 in cell-free conditions and examined the potential protein glutathionylation by immunoblotting. We found no significant difference in GAPDH glutathionylation level between oxidant-treated cells and controls (data not shown). Moreover, GAPDH oxidized in glutathione-free conditions (Figure 6, top) has an identical thiol pattern to GAPDH oxidized by cell treatment with oxidant (Figure 6, bottom). Little and O’Brien clearly demonstrated that H2O2 oxidizes GAPDH thiols via formation of sulfenic acid,20 and our experiments are consistent with formation of -S-O-H groups as the most likely mechanism of GAPDH thiol oxidation by OxLDL.
In view of the lack of the effect of OxLDL on GAPDH mRNA levels, we performed pulse-chase labeling experiments to assess the potential effect of OxLDL or H2O2 on GAPDH protein stability. As shown in Figure 7A and 7B, either OxLDL or H2O2 markedly decreased GAPDH protein half-life (to 18±2 and 4±1 hours, respectively, compared with >24 hours in controls).
Proteasome Inhibitors MG-132 and Lactacystin Do Not Inhibit GAPDH Downregulation In Vivo, but Oxidized GAPDH Has Increased Sensitivity to Proteasome-Mediated Degradation In Vitro
Because it has been shown that H2O2-mediated protein oxidation "marks" proteins for preferential utilization in protein-degradation pathways,21 we compared the sensitivity of the oxidized and nonoxidized GAPDH to proteasome-mediated degradation in vitro. GAPDH purified from OxLDL-treated HASMC was completely degraded after 2 hours of incubation in a proteasome protein-degradation assay; however, protein level of GAPDH from control cells was not changed. Similarly, H2O2-oxidized GAPDH or GAPDH purified from H2O2-exposed cells was degraded faster than control protein (Figure 7C).
However, interestingly, pretreatment of cells with the proteasome inhibitor MG-132 (10 μmol/L) or lactacystin (10 μmol/L) did not inhibit OxLDL-induced downregulation of GAPDH (Figure 7D), suggesting that alternative non–proteasome-dependent pathway(s) are primarily responsible for OxLDL-induced degradation of GAPDH protein in vivo.
OxLDL-Specific GAPDH Downregulation Decreases Cell Glycolysis Rate and ATP Stores
We assessed energy metabolism in LDL-treated cells by measurement of glycolysis rates and total ATP levels. Glucose usage was inhibited by 65±7% in HASMC treated with OxLDL (60 μg/mL, 16 hours) compared with cells incubated in lipid-free media, whereas incubation with nonoxidized lipoprotein slightly stimulated cell glycolysis (Figure 8A).
To determine whether downregulation of GAPDH expression sufficed to reduce glycolysis rates, we used GAPDH-specific siRNA to downregulate GAPDH. As shown in Figure 8D, 50 nmol/L anti-GAPDH siRNA markedly reduced HASMC glycolysis rate. In addition, total ATP levels were reduced by 82±4% in OxLDL-treated cells (Figure 8B) and by 89±6% in siRNA-treated cells (Figure 8E) compared with control. These data indicate that GAPDH downregulation results in dramatic reduction of cell glycolysis rates and ATP depletion.
Discussion
GAPDH plays a key role in energy metabolism by conversion of glyceraldehyde-3-phosphate to 1, 3-bisphospho-glycerate in the glycolytic pathway. GAPDH has been referred to as a housekeeping protein based on the classical view that GAPDH expression remains relatively constant under changing cellular conditions. However, GAPDH is now known to contribute to a number of cellular functions unrelated to glycolysis that include nuclear RNA export and DNA replication and repair, exocytotic membrane fusion, cytoskeletal organization and phosphotransferase activity, apoptosis, and viral pathogenesis.1
In the present study, we explored the potential effect of oxidative stress on GAPDH expression using a model with significant physiological relevance, namely, the effects of OxLDL, a critically important proatherogenic agent4–6,9,22 on one of its main cellular targets, namely vascular smooth muscle. We have previously shown that OxLDL is localized to vascular smooth muscle in both early and advanced human atherosclerotic plaques22 and that OxLDL downregulates expression of insulin-like growth factor-1 (IGF-1) and its receptor in vascular smooth muscle.6,7,8 Here, we report that OxLDL markedly time- and dose-dependently downregulates GAPDH expression in HASMC via a redox-dependent mechanism, leading to a reduction in protein stability. Under the same experimental conditions, protein levels of other housekeeping molecules, namely -actin and -tubulin (data not shown) were not changed.
To identify mechanisms mediating OxLDL downregulation of GAPDH, we measured ROS formation in LDL-treated HASMC and demonstrated OxLDL-specific increases in total peroxides, superoxides, and H2O2. Antioxidants (Trolox, Tiron), LOX inhibitor (AA-861), and anti-CD36 receptor antibody suppressed both OxLDL-specific increase in ROS and downregulation of GAPDH and vice versa: inhibitors with no effect on ROS (allopurinol, indomethacin, NS-398) failed to prevent GAPDH downregulation. These results strongly suggest that OxLDL-generated ROS are critically important in mediating the downregulation of GAPDH.
We have recently shown, using a proteomic screen, that the CD36 receptor is markedly upregulated in HASMC exposed to OxLDL.7 This receptor is responsible for the uptake of approximately 80% of modified LDL by smooth muscle cells.23 In the present study, we report that the CD36 receptor is essential for both OxLDL-induced ROS formation and GAPDH downregulation. Our data suggest that OxLDL binding to CD36 results in enhanced ROS generation by LOX/NADPH oxidase, consistent with a novel functional relationship between CD36 and oxidases to generate ROS in HASMC.
ROS may be generated by different intracellular sources. The membrane-bound NADPH oxidase is a predominant source of ROS in multiple pathophysiological conditions in vascular smooth muscle.14 For instance, NADPH oxidase plays an essential role in angiotensin II–specific ROS formation in vascular smooth muscle.24 5- and 12/15-LOX also contribute to generation of ROS in SMC and 5-LOX or 12/15-LOX deficiency suppressed atherosclerosis in apolipoprotein E–deficient mice.25,26 We demonstrated that specific inhibitors of NADPH oxidase assembly decreased OxLDL-specific ROS and partially prevented downregulation of GAPDH and that GAPDH downregulation was completely blocked in p47phox–/– vascular SMC. The LOX inhibitor AA-861 and 12/15-LOX deficiency also reduced OxLDL-induced ROS generation and GAPDH downregulation. These data strongly suggest that OxLDL uses at least 2 different oxidases (NADPH oxidase and 12/15-LOX) for generation of ROS and downregulation of GAPDH in response to OxLDL in vascular SMC.
Both total peroxide measurements with CDC-H2F and specific H2O2 detection by Amplex Red show that OxLDL quickly enhances cellular H2O2 level. Our previous observation that OxLDL compounds, oxidized fatty acids, could also increase intracellular levels of H2O2 in smooth muscle cells, mediating its cytotoxic effects,27 is consistent with these data. We report here that H2O2 scavengers, but not scavenging of superoxides, blocked the OxLDL-specific decrease in GAPDH. Furthermore, exogenous H2O2 mimics the effect of OxLDL and dose-dependently downregulates GAPDH. Taken together, these data indicate a major contribution of H2O2 in ROS-mediated GAPDH suppression.
We found that SODs have differential effects on OxLDL-specific GAPDH suppression: SOD acts as a partial blocker, but PEG-SOD is able to slightly potentiate GAPDH suppression. Dual functional roles of SOD (superoxide scavenger/H2O2 generator)18 could explain the difference between these effects. PEG-SOD, an enzyme with preferential intracellular distribution,28 decreases superoxides but also increases H2O2, which may reach toxic levels in some cellular compartments. In contrast, SOD-dependent H2O2 increase in the extracellular media may be less critical, because of dilution and/or antioxidant activity in conditioned medium. Our data suggest that one mechanism that could contribute to OxLDL-induced H2O2 elevation is the conversion of high levels of superoxides in OxLDL-treated cells into "extra" H2O2 by different active SODs present in HASMC. Our finding that OxLDL suppressed activity of the H2O2 scavenger, catalase, could be an additional mechanism contributing to the increase in cellular H2O2 levels.
Our data indicate that the primary mechanism for OxLDL downregulation of GAPDH is a reduction in protein stability, as shown by pulse-chase labeling experiments. OxLDL-induced GAPDH oxidation, and namely S-thiolation, was likely mediated via an increase in intracellular H2O2 (consistent with previous reports that GAPDH may be oxidized by exogenous H2O2 in endothelial cells19,29) and resulted in a major decrease in GAPDH protein half-life. Thus, oxidized GAPDH protein was less stable than nonoxidized protein, as evidenced by the increased sensitivity of OxLDL- or H2O2-oxidized GAPDH to proteasome system-mediated degradation. These data strongly suggest that OxLDL-induced H2O2-dependent GAPDH oxidation makes this protein a preferential substrate for cellular protein degradation.
We found that the proteasome inhibitor MG-132 efficiently blocks oxidized GAPDH degradation in vitro (Figure 7C, top); however, this inhibitor as well as lactacystin failed to prevent OxLDL-specific GAPDH downregulation in HASMC (Figure 7D). As reported recently, 4-hydroxy-2-nonenal–induced GAPDH degradation in a leukemia cell line was also resistant to pretreatment with MG-132,30 suggesting that a system other than the ubiquitin–proteasome pathway was responsible for GAPDH degradation in vivo. Because a variety of protein-degradation systems mediate removal of oxidized proteins from the cell, including the proteasome system,31 a complex process of chaperone-mediated autophagy (lysosomes)32 and endoplasmic reticulum-associated degradation,33 identification of specific system(s) that mediate oxidized GAPDH degradation in vivo is an important area for further investigation.
Great variability has been reported for ATP levels within the atherosclerotic arterial wall depending on lesion stage, development of necrosis, the presence of lipoproteins, oxygen availability, and local levels of glucose.34 Notably, it has been demonstrated that ATP levels were strongly decreased in atherosclerotic lesions that developed in combination with hypercholesterolemia.35 Consistent with this in vivo study, we report here that there is marked suppression of cell glucose usage and a reduction in levels of ATP in vascular smooth muscle cells exposed to a major proatherogenic molecule, OxLDL. Furthermore, downregulation of GAPDH using siRNA produced equivalent reductions in glycolysis rate and ATP levels. Our data strongly suggest that OxLDL-specific GAPDH downregulation results in inhibition of glycolysis, which, in turn, probably, mediates general ATP depletion. We have also found that OxLDL decreases levels of 1 of the major glycolysis products, L-lactate (S. Sukhanov, unpublished data, 2005). Our findings suggest that alterations in arterial wall energy homeostasis, related to OxLDL downregulation of GAPDH, may play an important role in mechanisms of atherogenesis.
Acknowledgments
We thank Dr T. Peng (Baylor College of Medicine, Houston, Tex) for the excellent assistance in preparation of LDL-treated cell lysates. We thank Dr R. Natarajan (Beckman Research Institute of the City of Hope, Calif) for kindly providing wild-type and 12/15-LOX–deficient cells and Drs M. Runge and N. Madamanchi (University of North Carolina at Chapel Hill, NC) for SMC derived from p47phox–/– mice.
Source of Funding
This study was supported by NIH grant HL70241.
Disclosures
None.
Footnotes
Original received January 14, 2005; resubmission received January 3, 2006; revised resubmission received May 20, 2006; accepted June 2, 2006.
References
Sirover MA. New nuclear functions of the glycolytic protein, glyceraldehyde-3-phosphate dehydrogenase, in mammalian cells. J Cell Biochem. 2005; 95: 45–52. [Order article via Infotrieve]
McKenzie SM, Doe WF, Buffinton GD. 5-Aminosalicylic acid prevents oxidant mediated damage of glyceraldehyde-3-phosphate dehydrogenase in colon epithelial cells. Gut. 1999; 44: 180–185.
Wentzel P, Ejdesjo A, Eriksson UJ. Maternal diabetes in vivo and high glucose in vitro diminish GAPDH activity in rat embryos. Diabetes. 2003; 52: 1222–1228.
Parthasarathy S, Santanam N, Ramachandran S, Meilhac O. Oxidants and antioxidants in atherogenesis. An appraisal. J Lipid Res. 1999; 40: 2143–2157.
Parthasarathy S, Rankin SM. Role of oxidized low density lipoprotein in atherogenesis. Prog Lipid Res. 1992; 31: 127–143. [Order article via Infotrieve]
Higashi Y, Peng T, Du J, Sukhanov S, Li Y, Itabe H, Parthasarathy S, Delafontaine P. A redox-sensitive pathway mediates oxidized LDL-induced downregulation of insulin-like growth factor-1 receptor. J Lipid Res. 2005; 46: 1266–1277.
Sukhanov S, Delafontaine P. Protein chip-based microarray profiling of oxidized low density lipoprotein-treated cells. Proteomics. 2005; 5: 1274–1280. [Order article via Infotrieve]
Sukhanov S, Song YH, Delafontaine P. Global analysis of differentially expressed genes in oxidized LDL-treated human aortic smooth muscle cells. Biochem Biophys Res Commun. 2003; 306: 443–449. [Order article via Infotrieve]
Li Y, Higashi Y, Itabe H, Song YH, Du J, Delafontaine P. Insulin-like growth factor-1 receptor activation inhibits oxidized LDL-induced cytochrome C release and apoptosis via the phosphatidylinositol 3 kinase/Akt signaling pathway. Arterioscler Thromb Vasc Biol. 2003; 23: 2178–2184.
Napoli C, de Nigris F, Palinski W. Multiple role of reactive oxygen species in the arterial wall. J Cell Biochem. 2001; 82: 674–682. [Order article via Infotrieve]
Nicholson AC, Han J, Febbraio M, Silversterin RL, Hajjar DP. Role of CD36, the macrophage class B scavenger receptor, in atherosclerosis. Ann N Y Acad Sci. 2001; 947: 224–228.
Jala VR, Haribabu B. Leukotrienes and atherosclerosis: new roles for old mediators. Trends Immunol. 2004; 25: 315–322. [Order article via Infotrieve]
Reddy MA, Kim YS, Lanting L, Natarajan R. Reduced growth factor responses in vascular smooth muscle cells derived from 12/15-lipoxygenase-deficient mice. Hypertension. 2003; 41: 1294–1300.
Cai H, Griendling KK, Harrison DG. The vascular NAD(P)H oxidases as therapeutic targets in cardiovascular diseases. Trends Pharmacol Sci. 2003; 24: 471–478. [Order article via Infotrieve]
Barry-Lane PA, Patterson C, van der Merwe M, Hu Z, Holland SM, Yeh ET, Runge MS. p47phox is required for atherosclerotic lesion progression in ApoE(-/-) mice. J Clin Invest. 2001; 108: 1513–1522.
Watanabe T, Pakala R, Katagiri T, Benedict CR. Oxidized low-density lipoproteins potentiate the mitogenic effect of 5-hydroxytryptamine on vascular smooth muscle cells. Jpn Heart J. 2002; 43: 35–42. [Order article via Infotrieve]
Suematsu M, Schmid-Schonbein GW, Chavez-Chavez RH, Yee TT, Tamatani T, Miyasaka M, Delano FA, Zweifach BW. In vivo visualization of oxidative changes in microvessels during neutrophil activation. Am J Physiol. 1993; 264: H881–H891. [Order article via Infotrieve]
Faraci FM, Didion SP. Vascular protection: superoxide dismutase isoforms in the vessel wall. Arterioscler Thromb Vasc Biol. 2004; 24: 1367–1373.
Schuppe-Koistinen I, Moldeus P, Bergman T, Cotgreave IA. S-thiolation of human endothelial cell glyceraldehyde-3-phosphate dehydrogenase after hydrogen peroxide treatment. Eur J Biochem. 1994; 221: 1033–1037.
Little C, O’Brien PJ. Mechanism of peroxide-inactivation of the sulphydryl enzyme glyceraldehyde-3-phosphate dehydrogenase. Eur J Biochem. 1969; 10: 533–538. [Order article via Infotrieve]
Stadtman ER. Metal ion-catalyzed oxidation of proteins: biochemical mechanism and biological consequences. Free Radic Biol Med. 1990; 9: 315–325. [Order article via Infotrieve]
Okura Y, Brink M, Itabe H, Scheidegger KJ, Kalangos A, Delafontaine P. Oxidized low-density lipoprotein is associated with apoptosis of vascular smooth muscle cells in human atherosclerotic plaques. Circulation. 2000; 102: 2680–2686.
Sukhanov STJ, Serrano C, Bowers C, Delafontaine P. Anti-atherogenic effects of growth hormone releasing peptide-2 in vitro: potential implications in prevention of atherosclerosis. American Heart Association Scientific Sessions, November 13–16, 2005, Dallas, Tex; 2005:651 Abstract.
Griendling KK, Minieri CA, Ollerenshaw JD, Alexander RW. Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells. Circ Res. 1994; 74: 1141–1148.
Mehrabian M, Allayee H, Wong J, Shi W, Wang XP, Shaposhnik Z, Funk CD, Lusis AJ. Identification of 5-lipoxygenase as a major gene contributing to atherosclerosis susceptibility in mice. Circ Res. 2002; 91: 120–126.
Huo Y, Zhao L, Hyman MC, Shashkin P, Harry BL, Burcin T, Forlow SB, Stark MA, Smith DF, Clarke S, Srinivasan S, Hedrick CC, Pratico D, Witztum JL, Nadler JL, Funk CD, Ley K. Critical role of macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 110: 2024–2031.
Santanam N, Aug N, Zhou M, Keshava C, Parthasarathy S. Overexpression of human catalase gene decreases oxidized lipid-induced cytotoxicity in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1999; 19: 1912–1917.
Muelleman RL, Grandstaff PM, Robinson WA. The use of pegorgotein in the treatment of frostbite. Wilderness Environ Med. 1997; 8: 17–19. [Order article via Infotrieve]
Schuppe-Koistinen I, Gerdes R, Moldeus P, Cotgreave IA. Studies on the reversibility of protein S-thiolation in human endothelial cells. Arch Biochem Biophys. 1994; 315: 226–234. [Order article via Infotrieve]
Tsuchiya Y, Yamaguchi M, Chikuma T, Hojo H. Degradation of glyceraldehyde-3-phosphate dehydrogenase triggered by 4-hydroxy-2-nonenal and 4-hydroxy-2-hexenal. Arch Biochem Biophys. 2005; 438: 217–222. [Order article via Infotrieve]
Divald A, Powell SR. Proteasome mediates removal of proteins oxidized during myocardial ischemia. Free Radic Biol Med. 2006; 40: 156–164. [Order article via Infotrieve]
Kiffin R, Christian C, Knecht E, Cuervo AM. Activation of chaperone-mediated autophagy during oxidative stress. Mol Biol Cell. 2004; 15: 4829–4840.
Molteni SN, Fassio A, Ciriolo MR, Filomeni G, Pasqualetto E, Fagioli C, Sitia R. Glutathione limits Ero1-dependent oxidation in the endoplasmic reticulum. J Biol Chem. 2004; 279: 32667–32673.
Levin M, Leppanen O, Evaldsson M, Wiklund O, Bondjers G, Bjornheden T. Mapping of ATP, glucose, glycogen, and lactate concentrations within the arterial wall. Arterioscler Thromb Vasc Biol. 2003; 23: 1801–1807.
Heinle H. Metabolite concentration gradients in the arterial wall of experimental atherosclerosis. Exp Mol Pathol. 1987; 46: 312–320. [Order article via Infotrieve]
the Cardiology Section (S.S., Y.H., S.-Y.S, P.D.), Department of Medicine, Tulane University, New Orleans, La
Department of Biological Chemistry (H.I.), Showa University, Tokyo, Japan
Innoshima City Medical Associated Hospital (K.O.), Japan
Pathology Department (S.P.), Louisiana State University, New Orleans.
Abstract
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is a classical glycolytic enzyme that is involved in cellular energy production and has important housekeeping functions. We used the natural prooxidant and proatherogenic molecule oxidized low-density lipoprotein (OxLDL) to determine a potential link between OxLDL-promoted oxidative stress, GAPDH expression, and smooth muscle cell energy metabolism. OxLDL but not native LDL (nLDL) produced a 60% to 100% dose- and time-dependent reduction of GAPDH protein. OxLDL increased reactive oxygen species (ROS) formation, including rapid elevation of H2O2 levels. OxLDL decreased intracellular catalase expression, likely contributing to the increase in H2O2. Antioxidants, anti-CD36 receptor antibody, NADPH oxidase, or lipoxygenase blockers decreased OxLDL-specific ROS and prevented GAPDH downregulation. 12/15-Lipoxygenase or p47phox deficiency resulted in attenuation of GAPDH downregulation, but 5-lipoxygenase suppression had no effect. OxLDL or exogenous H2O2 oxidized GAPDH thiols, decreasing GAPDH protein half-life and increasing GAPDH sensitivity to proteasome-mediated protein degradation in vitro. OxLDL- or small interfering RNA–specific downregulation of GAPDH resulted in 65% reduction in glycolysis rate and 82% decrease in ATP levels. In conclusion, our data demonstrate that OxLDL downregulated GAPDH via a H2O2-dependent decrease in protein stability. GAPDH protein damage resulted in marked depletion of cellular ATP levels. Our data have important implications for understanding the metabolic effect of OxLDL on the vessel wall and mechanism of atherogenesis.
Key Words: reactive oxygen species low-density lipoprotein energy metabolism vascular smooth muscle
Introduction
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (EC 1.2.1.12) is a glycolytic enzyme involved in energy production and is one of the most abundant cellular proteins. GAPDH is constitutively expressed in different cells and is viewed as an important housekeeping molecule. However, it has become clear that the role of GAPDH is not restricted to its pivotal glycolytic function. Indeed, GAPDH regulates multiple other processes, including membrane fusion, nuclear RNA transport, and DNA replication and repair.1
Increasing evidence suggests that GAPDH is a preferential target for oxidative stress. Nitric oxide (NO) and reactive oxygen species (ROS)-mediated oxidation of GAPDH thiols (S-thiolation) and subsequent inhibition of enzyme activity have been described in colon epithelial cells during inflammatory bowel disease.2 Wentzel et al3 demonstrated that diabetic embryopathy in the rat is associated with the inhibition of GAPDH activity and a decrease in GAPDH mRNA and that this effect is blunted by the antioxidant N-acetylcysteine. Taken together, these observations indicate a high sensitivity of GAPDH to cell redox status and potential involvement of this enzyme in oxidative stress-promoted events.
Oxidized low-density lipoprotein (OxLDL) is a powerful natural prooxidant derived in vivo from nLDL by cell-mediated oxidation.4 OxLDL is a key proatherogenic agent5 that induces, as we have shown, an increase in ROS formation, specific alterations in gene/protein expression, and apoptosis of vascular smooth muscle cells.6–9 We designed this study to determine a potential link between OxLDL-specific increase in ROS, GAPDH expression, and human aortic smooth muscle cell (HASMC) energy metabolism. Our findings indicate dramatic OxLDL-induced downregulation of GAPDH via a redox-dependent signal pathway, resulting in depletion of cellular energy stores. These results provide a potential mechanism for understanding metabolic alterations of the vasculature during the process of atherogenesis.
Materials and Methods
Cell culture, isolation, oxidative modification of LDL, and immunoblotting analysis protocols have been described previously.6 Formation of intracellular ROS were assessed with 6-carboxy-2',7'-dichlorodihydro-fluorescein diacetate (CDC-H2F), hydroethidine, and Amplex Red kit (all from Molecular Probes). GAPDH free thiol status was measured by labeling of cell extracts with the thiol-sensitive fluorescent dye fluorescein-5-maleimide (Molecular Probes), and metabolic labeling of HASMC was performed with EXPRE35S35S Protein Labeling Mix (Perkin Elmer) followed by immunoprecipitation with anti-GAPDH (Chemicon) or anti–-actin (Sigma) antibody. An expanded Materials and Methods section containing details regarding ROS measurements, GAPDH purification, proteasome protein-degradation assay, and use of small interfering RNA (siRNA), as well as Figures I through III, are given in the data supplement, available at http://circres.ahajournals.org.
Results
OxLDL Downregulates GAPDH Protein
Confluent (80% to 90%) HASMC were treated with 60 μg/mL OxLDL or native LDL (nLDL) (control) for 0 to 24 hours or incubated in serum-free media (SFM). No change in GAPDH expression was found after cell treatment with nLDL or SFM; in contrast, cell incubation with OxLDL time-dependently downregulated GAPDH protein (Figure 1A). GAPDH protein level was reduced in OxLDL-treated cells to 68±15% and 33±10% compared with control at 9 and 15 hours, respectively, and GAPDH expression at 24 hours was virtually not detectable.
HASMC treatment by LDLs for 15 hours was selected for generation of dose-response curves and to assess effects of antioxidants. For dose-response experiments, HASMC were treated with 0 to 100 μg/mL OxLDL or nLDL and GAPDH expression was analyzed by immunoblotting. Sixty micrograms per milliliter OxLDL decreased GAPDH levels to 38±11%, and 80 μg/mL OxLDL downregulated GAPDH to 25±3%. OxLDL (20 to 100 μg/mL) produced no changes in protein levels of other housekeeping proteins -actin and -tubulin (Figure 1B). However, we noted that treatment of HASMC with highly OxLDL (>50 nmol malondialdehyde per milligram protein) slightly decreased -actin levels (15±5% compared with nLDL-treated cells), as evidenced by immunoblotting, together with marked downregulation of GAPDH protein levels.
Immunoblotting data obtained with mouse monoclonal anti-GAPDH antibody were further confirmed with rabbit polyclonal anti-GAPDH antibody (data not shown). We also examined the efficiency of the recognition of potentially oxidized antigen (GAPDH) by cognitive antibody (mouse anti-GAPDH) in a routine Western analysis. Purified chicken muscle GAPDH (0.1 mg/mL; Sigma) was directly oxidized by 1.5 mmol/L H2O2, and an aliquot of oxidized and untreated GAPDH (control) were analyzed by immunoblotting. No differences were found between band density of oxidized GAPDH and control (data not shown). Although the possibility remains that there is a loss of immunoreactivity independent of reduced expression of the GAPDH protein. This appears unlikely in view of our results comparing immunoreactivity of oxidized and nonoxidized GAPDH using 2 different epitope-specific anti-GAPDH antibodies.
GAPDH gene expression was analyzed by Northern blotting. In time-course experiments, GAPDH protein was decreased after 9 hours of OxLDL treatment, and for this period (0 to 9 hours), GAPDH mRNA levels were not changed significantly compared with nLDL (Figure 1C). However, longer incubation with OxLDL (15 to 24 hours) produced a slight decrease of GAPDH mRNA (18±5% at 24 hours).
Antioxidants Suppress OxLDL-Specific Peroxides, Decrease Superoxides, and Prevent Downregulation of GAPDH
Because OxLDL signaling can involve ROS,10 we assessed formation of intracellular peroxides and superoxides in LDL-treated HASMC with the peroxide-sensitive dye CDC-H2F and the superoxide-specific dye hydroethidine (HE). Both oxidized and native LDL (60 μg/mL) increased cell peroxides time dependently up to 9 hours of treatment. For 2 to 6 hours of incubation, OxLDL produced 40% to 70% more peroxides than nLDL, but longer treatments resulted in no difference between peroxide levels induced by OxLDL and control (see supplemental Figure IA). Unlike peroxide formation, superoxide level was significantly higher in cells treated with OxLDL versus nLDL for all time points up to 30 hours. OxLDL increased peroxides dose dependently (20 to 80 μg/mL), with a peak formation at 3 to 4 hours. OxLDL-specific formation of superoxides was also dose-dependent with a maximum at 20 hours (supplemental Figure IB and ID). These time points were chosen for assaying OxLDL-specific peroxide and superoxide generation in further experiments.
Pretreatment with the water-soluble vitamin E derivative Trolox (0.25 mmol/L) or antioxidant Tiron (1 mmol/L) dramatically reduced basal and nLDL-induced peroxides, with complete elimination of OxLDL-specific peroxides (Figure 2A). In contrast, antioxidants were less efficient in inhibition of cell superoxides induced by OxLDL; both nLDL- or OxLDL-induced superoxides were equally suppressed by antioxidants (Figure 2B). Trolox or Tiron completely prevented OxLDL-specific decrease in GAPDH (Figure 2C). Based on this data, we concluded that ROS mediate the OxLDL-specific GAPDH decrease.
CD36 Scavenger Receptor, 12/15-Lipoxygenase, and NADPH Oxidase Mediate OxLDL-Induced ROS Increase and GAPDH Downregulation
The CD36 scavenger receptor mediates the uptake of OxLDL, allowing OxLDL-specific intracellular signaling and development of atherosclerosis.11 Using real time PCR we confirmed the presence of CD36 mRNA in HASMC and one unique protein band immunogenic to CD36 antibody (MW=90±7 kDa, n=3) was detected in cell lysates by immunoblotting (data not shown). To verify the potential involvement of CD36 receptor in OxLDL-specific ROS increase and GAPDH downregulation we used mouse FA6–152 anti-CD36 blocking antibody (Immunotech, Cedex, France), a specific inhibitor of CD36-dependent OxLDL uptake.11 We found that anti-CD36 antibody (20 μg/mL) decreased OxLDL-specific increase in peroxide and superoxide formation, whereas pretreatment with nonspecific antibody (20 μg/mL normal mouse IgG) had no effect (see supplemental Figure II). Antibody-specific inhibition of CD36 scavenger receptor resulted in complete inhibition of OxLDL-induced GAPDH decrease (Figure 3A).
We used allopurinol, the xanthine oxidase inhibitor; NS-398, the specific blocker of cyclooxygenase-2; indomethacin, the nonspecific blocker of cyclooxygenase-1 and cyclooxygenase-2; and rotenone, the inhibitor of mitochondrial complex I, to study potential involvement of these ROS producing systems in OxLDL-dependent increase in HASMC peroxides and superoxides. Preincubation of HASMC with 100 μmol/L allopurinol, 10 μmol/L NS-398, or 10 μmol/L indomethacin did not block OxLDL-specific peroxide or superoxide formation or downregulation of GAPDH, and rotenone was toxic at concentrations of 10 and 100 μmol/L (data not shown).
Lipoxygenases (LOXs), such as 5-LOX, which is a key enzyme in leukotriene production, and 12/15-LOX are a family of lipid peroxidizing enzymes that are another source of ROS in the vasculature.12 We verified the presence of 5- and 12/15-LOX proteins and corresponding enzyme activities in HASMC (data not shown). We found that the LOX inhibitor AA-861 (5 μmol/L) markedly reduced OxLDL-induced ROS production (supplemental Figure II). The same dose of AA-861 completely blocked OxLDL downregulation of GAPDH (Figure 3B). To identify the LOX form that mediates OxLDL-induced ROS production and GAPDH downregulation, we used 5-LOX–specific and nonspecific (control) siRNA-treated HASMC and vascular smooth muscle cells (SMC) obtained from 12/15-LOX–deficient mice13 and wild-type mice (control). All methods involving animals were approved by the Institutional Animal Care and Use Committee (IACUC) of Tulane University Health Sciences Center. siRNA-mediated 5-LOX protein decrease was 65% as measured by immunoblotting analysis (Figure 3D), and there was a similar decrease in 5-LOX mRNA levels as measured by real-time PCR (data not shown). We found that 12/15-LOX deficiency resulted in a 62±7% reduction of OxLDL-induced peroxides; however, 5-LOX suppression had no effect (supplemental Figure III). OxLDL-induced downregulation of GAPDH was significantly blunted in 12/15-LOX–/– vascular SMC (54±14% inhibition compared with wild-type), whereas 5-LOX suppression had no blocking effect on OxLDL-induced GAPDH downregulation (Figure 3E).
The membrane-bound NADPH oxidase is a predominant source of ROS in multiple pathophysiological conditions in vascular smooth muscle.14 We found that apocynin (50 μmol/L), a potent and selective inhibitor of NADPH oxidase, and the structurally unrelated NADPH oxidase blocker AEBSF (1 mmol/L) were less efficient in suppression of OxLDL-specific ROS compared with AA-861 (supplemental Figure II). NADPH oxidase blockers only partially suppressed downregulation of GAPDH (Figure 3C). p47phox is a critical subunit of the NADPH oxidase, and a null mutation of this gene resulted in attenuation of superoxide production in SMC and suppression of atherosclerosis.15 To verify our data obtained with NADPH oxidase inhibitors, we measured peroxides and GAPDH protein levels in vascular SMC from p47phox–/– or wild-type (control) mice exposed to nLDL or OxLDL. We found that p47phox–/– cells produce 82% less peroxides in response to OxLDL compared with wild type (supplemental Figure III). p47phox deficiency completely blocked OxLDL-induced GAPDH downregulation (Figure 3F).
These data suggest involvement of the CD36 receptor, 12/15-LOX, and NADPH oxidase in ROS formation and OxLDL-specific GAPDH downregulation.
H2O2 Scavengers Completely Block OxLDL-Specific GAPDH Downregulation, and Exogenous H2O2 Downregulates GAPDH
H2O2 is a major oxidative component of OxLDL-induced ROS16 and, therefore, could be involved in ROS-mediated GAPDH downregulation. We measured specific H2O2 release from LDL-exposed cells with/without pretreatment with Trolox (0.25 mmol/L), specific H2O2 scavengers catalase (10 kU/mL), or polyethylene glycol–coupled catalase (PEG-catalase) (1 kU/mL) using the Amplex Red assay. OxLDL (60 μg/mL, 3 hours) significantly increased H2O2 levels compared with nLDL or SFM (Figure 4A). Pretreatment with antioxidant or catalase efficiently suppressed OxLDL-induced H2O2 formation to below basal levels. After calibration of the Amplex Red signal with standards (H2O2, 0 to 1000 μmol/L), we estimated that the OxLDL-specific increase in cellular H2O2 level corresponds to 350 to 450 μmol/L H2O2 (data not shown). This is consistent with reports showing that oxidative stress enhanced cellular H2O2 concentration in vivo to the submillimolar range.17 Therefore, we examined GAPDH expression in response to similar doses of exogenous H2O2. We found that HASMC incubation with H2O2 (225 to 450 μmol/L, 16 hours) resulted in a downregulation of GAPDH and that this effect was dose dependent (Figure 4B).
Based on these data, we hypothesized that OxLDL-specific increase in H2O2 is critical for subsequent GAPDH protein suppression and verified this hypothesis using H2O2 scavengers (catalase and PEG-catalase) and superoxide scavengers (superoxide dismutase [SOD] and PEG-SOD). Only catalases completely blocked OxLDL-induced downregulation of GAPDH (Figure 4C and 4D). The blocking effect of SOD was partial: GAPDH levels still decreased in OxLDL-treated cells compared with nLDL (67±7% of control, n=3, P<0.05) after preincubation with this superoxide scavenger. PEG-SOD at 50 U/mL even potentiated GAPDH downregulation and was toxic at higher concentrations. This data suggested that OxLDL-specific increase in H2O2 is responsible for GAPDH downregulation.
OxLDL Rapidly Decreases Catalase Activity in HASMC and Does Not Alter SOD Activity
Two major antioxidant enzyme systems protect cells from oxidative stress by maintaining safe levels of ROS: SODs, enzymes that catalyze conversion of superoxides to H2O2 including intracellular (Mn-SOD, Cu/Zn-SOD) and extracellular (EC-SOD) forms and catalase.18 To clarify the potential mechanism of OxLDL-induced H2O2 elevation, we measured the activity and protein levels of SODs and catalase in HASMC treated by OxLDL.
Immunoblotting analysis showed that the relative amount of catalase was decreased by 58±5% by OxLDL treatment compared with nLDL, but Mn-SOD protein level was not changed (Figure 5A and 5B). Similarly, the functional activity of catalase was significantly reduced (Figure 5C). However, activity of intracellular SOD, as assessed by Cayman Chemical assay and in situ SOD activity detection, was not different between OxLDL- and nLDL-treated cells (data not shown). Additionally, Western blotting revealed that OxLDL did not alter expression of Cu/Zn-SOD (data not shown). We detected activity of an EC-SOD–like protein in conditioned media from LDL-treated cells using in situ assay and found that this activity also was unregulated by OxLDL (data not shown). These data suggest that the reduction in catalase activity and the maintenance of SOD activity could contribute to the increase in intracellular H2O2 in response to OxLDL.
OxLDL or H2O2 Oxidize GAPDH Thiols and Decrease Protein Stability
The GAPDH molecule has 3 easily oxidized SH groups and 2 of them (Cys149 and Cys153) are essential for GAPDH catalytic activity.19 We measured GAPDH thiol status following cell exposure to OxLDL or H2O2. GAPDH-labeling efficiency with the thiol-sensitive dye fluorescein-5-maleimide was dramatically decreased following treatment with OxLDL (60 μg/mL, 3 hours) or, similarly, after incubation with H2O2 (400 μmol/L, 3 hours) without alteration of GAPDH protein levels (Figure 6).
In theory, GAPDH thiol (R-S-H) modification by an oxidant can include: (1) formation of inter- or intramolecular disulphide bands (R-S-S-R); (2) formation of mixed disulphides with a small intracellular thiol, glutathione (R-S-S-R2); or (3) direct oxidation of the Cys thiol groups to sulfenic acid (R-S-O-H). To address the first possibility, we used nonreducing gel electrophoresis to examine whether formation of an intramolecular disulphide bond or a bond between 2 molecules of GAPDH resulted in altered protein migration by nonreducing SDS-PAGE. Neither H2O2 nor OxLDL treatment of cells resulted in altered GAPDH electrophoretic mobility (data not shown). To address the second possibility, we examined GAPDH thiol oxidation by H2O2 in cell-free conditions and examined the potential protein glutathionylation by immunoblotting. We found no significant difference in GAPDH glutathionylation level between oxidant-treated cells and controls (data not shown). Moreover, GAPDH oxidized in glutathione-free conditions (Figure 6, top) has an identical thiol pattern to GAPDH oxidized by cell treatment with oxidant (Figure 6, bottom). Little and O’Brien clearly demonstrated that H2O2 oxidizes GAPDH thiols via formation of sulfenic acid,20 and our experiments are consistent with formation of -S-O-H groups as the most likely mechanism of GAPDH thiol oxidation by OxLDL.
In view of the lack of the effect of OxLDL on GAPDH mRNA levels, we performed pulse-chase labeling experiments to assess the potential effect of OxLDL or H2O2 on GAPDH protein stability. As shown in Figure 7A and 7B, either OxLDL or H2O2 markedly decreased GAPDH protein half-life (to 18±2 and 4±1 hours, respectively, compared with >24 hours in controls).
Proteasome Inhibitors MG-132 and Lactacystin Do Not Inhibit GAPDH Downregulation In Vivo, but Oxidized GAPDH Has Increased Sensitivity to Proteasome-Mediated Degradation In Vitro
Because it has been shown that H2O2-mediated protein oxidation "marks" proteins for preferential utilization in protein-degradation pathways,21 we compared the sensitivity of the oxidized and nonoxidized GAPDH to proteasome-mediated degradation in vitro. GAPDH purified from OxLDL-treated HASMC was completely degraded after 2 hours of incubation in a proteasome protein-degradation assay; however, protein level of GAPDH from control cells was not changed. Similarly, H2O2-oxidized GAPDH or GAPDH purified from H2O2-exposed cells was degraded faster than control protein (Figure 7C).
However, interestingly, pretreatment of cells with the proteasome inhibitor MG-132 (10 μmol/L) or lactacystin (10 μmol/L) did not inhibit OxLDL-induced downregulation of GAPDH (Figure 7D), suggesting that alternative non–proteasome-dependent pathway(s) are primarily responsible for OxLDL-induced degradation of GAPDH protein in vivo.
OxLDL-Specific GAPDH Downregulation Decreases Cell Glycolysis Rate and ATP Stores
We assessed energy metabolism in LDL-treated cells by measurement of glycolysis rates and total ATP levels. Glucose usage was inhibited by 65±7% in HASMC treated with OxLDL (60 μg/mL, 16 hours) compared with cells incubated in lipid-free media, whereas incubation with nonoxidized lipoprotein slightly stimulated cell glycolysis (Figure 8A).
To determine whether downregulation of GAPDH expression sufficed to reduce glycolysis rates, we used GAPDH-specific siRNA to downregulate GAPDH. As shown in Figure 8D, 50 nmol/L anti-GAPDH siRNA markedly reduced HASMC glycolysis rate. In addition, total ATP levels were reduced by 82±4% in OxLDL-treated cells (Figure 8B) and by 89±6% in siRNA-treated cells (Figure 8E) compared with control. These data indicate that GAPDH downregulation results in dramatic reduction of cell glycolysis rates and ATP depletion.
Discussion
GAPDH plays a key role in energy metabolism by conversion of glyceraldehyde-3-phosphate to 1, 3-bisphospho-glycerate in the glycolytic pathway. GAPDH has been referred to as a housekeeping protein based on the classical view that GAPDH expression remains relatively constant under changing cellular conditions. However, GAPDH is now known to contribute to a number of cellular functions unrelated to glycolysis that include nuclear RNA export and DNA replication and repair, exocytotic membrane fusion, cytoskeletal organization and phosphotransferase activity, apoptosis, and viral pathogenesis.1
In the present study, we explored the potential effect of oxidative stress on GAPDH expression using a model with significant physiological relevance, namely, the effects of OxLDL, a critically important proatherogenic agent4–6,9,22 on one of its main cellular targets, namely vascular smooth muscle. We have previously shown that OxLDL is localized to vascular smooth muscle in both early and advanced human atherosclerotic plaques22 and that OxLDL downregulates expression of insulin-like growth factor-1 (IGF-1) and its receptor in vascular smooth muscle.6,7,8 Here, we report that OxLDL markedly time- and dose-dependently downregulates GAPDH expression in HASMC via a redox-dependent mechanism, leading to a reduction in protein stability. Under the same experimental conditions, protein levels of other housekeeping molecules, namely -actin and -tubulin (data not shown) were not changed.
To identify mechanisms mediating OxLDL downregulation of GAPDH, we measured ROS formation in LDL-treated HASMC and demonstrated OxLDL-specific increases in total peroxides, superoxides, and H2O2. Antioxidants (Trolox, Tiron), LOX inhibitor (AA-861), and anti-CD36 receptor antibody suppressed both OxLDL-specific increase in ROS and downregulation of GAPDH and vice versa: inhibitors with no effect on ROS (allopurinol, indomethacin, NS-398) failed to prevent GAPDH downregulation. These results strongly suggest that OxLDL-generated ROS are critically important in mediating the downregulation of GAPDH.
We have recently shown, using a proteomic screen, that the CD36 receptor is markedly upregulated in HASMC exposed to OxLDL.7 This receptor is responsible for the uptake of approximately 80% of modified LDL by smooth muscle cells.23 In the present study, we report that the CD36 receptor is essential for both OxLDL-induced ROS formation and GAPDH downregulation. Our data suggest that OxLDL binding to CD36 results in enhanced ROS generation by LOX/NADPH oxidase, consistent with a novel functional relationship between CD36 and oxidases to generate ROS in HASMC.
ROS may be generated by different intracellular sources. The membrane-bound NADPH oxidase is a predominant source of ROS in multiple pathophysiological conditions in vascular smooth muscle.14 For instance, NADPH oxidase plays an essential role in angiotensin II–specific ROS formation in vascular smooth muscle.24 5- and 12/15-LOX also contribute to generation of ROS in SMC and 5-LOX or 12/15-LOX deficiency suppressed atherosclerosis in apolipoprotein E–deficient mice.25,26 We demonstrated that specific inhibitors of NADPH oxidase assembly decreased OxLDL-specific ROS and partially prevented downregulation of GAPDH and that GAPDH downregulation was completely blocked in p47phox–/– vascular SMC. The LOX inhibitor AA-861 and 12/15-LOX deficiency also reduced OxLDL-induced ROS generation and GAPDH downregulation. These data strongly suggest that OxLDL uses at least 2 different oxidases (NADPH oxidase and 12/15-LOX) for generation of ROS and downregulation of GAPDH in response to OxLDL in vascular SMC.
Both total peroxide measurements with CDC-H2F and specific H2O2 detection by Amplex Red show that OxLDL quickly enhances cellular H2O2 level. Our previous observation that OxLDL compounds, oxidized fatty acids, could also increase intracellular levels of H2O2 in smooth muscle cells, mediating its cytotoxic effects,27 is consistent with these data. We report here that H2O2 scavengers, but not scavenging of superoxides, blocked the OxLDL-specific decrease in GAPDH. Furthermore, exogenous H2O2 mimics the effect of OxLDL and dose-dependently downregulates GAPDH. Taken together, these data indicate a major contribution of H2O2 in ROS-mediated GAPDH suppression.
We found that SODs have differential effects on OxLDL-specific GAPDH suppression: SOD acts as a partial blocker, but PEG-SOD is able to slightly potentiate GAPDH suppression. Dual functional roles of SOD (superoxide scavenger/H2O2 generator)18 could explain the difference between these effects. PEG-SOD, an enzyme with preferential intracellular distribution,28 decreases superoxides but also increases H2O2, which may reach toxic levels in some cellular compartments. In contrast, SOD-dependent H2O2 increase in the extracellular media may be less critical, because of dilution and/or antioxidant activity in conditioned medium. Our data suggest that one mechanism that could contribute to OxLDL-induced H2O2 elevation is the conversion of high levels of superoxides in OxLDL-treated cells into "extra" H2O2 by different active SODs present in HASMC. Our finding that OxLDL suppressed activity of the H2O2 scavenger, catalase, could be an additional mechanism contributing to the increase in cellular H2O2 levels.
Our data indicate that the primary mechanism for OxLDL downregulation of GAPDH is a reduction in protein stability, as shown by pulse-chase labeling experiments. OxLDL-induced GAPDH oxidation, and namely S-thiolation, was likely mediated via an increase in intracellular H2O2 (consistent with previous reports that GAPDH may be oxidized by exogenous H2O2 in endothelial cells19,29) and resulted in a major decrease in GAPDH protein half-life. Thus, oxidized GAPDH protein was less stable than nonoxidized protein, as evidenced by the increased sensitivity of OxLDL- or H2O2-oxidized GAPDH to proteasome system-mediated degradation. These data strongly suggest that OxLDL-induced H2O2-dependent GAPDH oxidation makes this protein a preferential substrate for cellular protein degradation.
We found that the proteasome inhibitor MG-132 efficiently blocks oxidized GAPDH degradation in vitro (Figure 7C, top); however, this inhibitor as well as lactacystin failed to prevent OxLDL-specific GAPDH downregulation in HASMC (Figure 7D). As reported recently, 4-hydroxy-2-nonenal–induced GAPDH degradation in a leukemia cell line was also resistant to pretreatment with MG-132,30 suggesting that a system other than the ubiquitin–proteasome pathway was responsible for GAPDH degradation in vivo. Because a variety of protein-degradation systems mediate removal of oxidized proteins from the cell, including the proteasome system,31 a complex process of chaperone-mediated autophagy (lysosomes)32 and endoplasmic reticulum-associated degradation,33 identification of specific system(s) that mediate oxidized GAPDH degradation in vivo is an important area for further investigation.
Great variability has been reported for ATP levels within the atherosclerotic arterial wall depending on lesion stage, development of necrosis, the presence of lipoproteins, oxygen availability, and local levels of glucose.34 Notably, it has been demonstrated that ATP levels were strongly decreased in atherosclerotic lesions that developed in combination with hypercholesterolemia.35 Consistent with this in vivo study, we report here that there is marked suppression of cell glucose usage and a reduction in levels of ATP in vascular smooth muscle cells exposed to a major proatherogenic molecule, OxLDL. Furthermore, downregulation of GAPDH using siRNA produced equivalent reductions in glycolysis rate and ATP levels. Our data strongly suggest that OxLDL-specific GAPDH downregulation results in inhibition of glycolysis, which, in turn, probably, mediates general ATP depletion. We have also found that OxLDL decreases levels of 1 of the major glycolysis products, L-lactate (S. Sukhanov, unpublished data, 2005). Our findings suggest that alterations in arterial wall energy homeostasis, related to OxLDL downregulation of GAPDH, may play an important role in mechanisms of atherogenesis.
Acknowledgments
We thank Dr T. Peng (Baylor College of Medicine, Houston, Tex) for the excellent assistance in preparation of LDL-treated cell lysates. We thank Dr R. Natarajan (Beckman Research Institute of the City of Hope, Calif) for kindly providing wild-type and 12/15-LOX–deficient cells and Drs M. Runge and N. Madamanchi (University of North Carolina at Chapel Hill, NC) for SMC derived from p47phox–/– mice.
Source of Funding
This study was supported by NIH grant HL70241.
Disclosures
None.
Footnotes
Original received January 14, 2005; resubmission received January 3, 2006; revised resubmission received May 20, 2006; accepted June 2, 2006.
References
Sirover MA. New nuclear functions of the glycolytic protein, glyceraldehyde-3-phosphate dehydrogenase, in mammalian cells. J Cell Biochem. 2005; 95: 45–52. [Order article via Infotrieve]
McKenzie SM, Doe WF, Buffinton GD. 5-Aminosalicylic acid prevents oxidant mediated damage of glyceraldehyde-3-phosphate dehydrogenase in colon epithelial cells. Gut. 1999; 44: 180–185.
Wentzel P, Ejdesjo A, Eriksson UJ. Maternal diabetes in vivo and high glucose in vitro diminish GAPDH activity in rat embryos. Diabetes. 2003; 52: 1222–1228.
Parthasarathy S, Santanam N, Ramachandran S, Meilhac O. Oxidants and antioxidants in atherogenesis. An appraisal. J Lipid Res. 1999; 40: 2143–2157.
Parthasarathy S, Rankin SM. Role of oxidized low density lipoprotein in atherogenesis. Prog Lipid Res. 1992; 31: 127–143. [Order article via Infotrieve]
Higashi Y, Peng T, Du J, Sukhanov S, Li Y, Itabe H, Parthasarathy S, Delafontaine P. A redox-sensitive pathway mediates oxidized LDL-induced downregulation of insulin-like growth factor-1 receptor. J Lipid Res. 2005; 46: 1266–1277.
Sukhanov S, Delafontaine P. Protein chip-based microarray profiling of oxidized low density lipoprotein-treated cells. Proteomics. 2005; 5: 1274–1280. [Order article via Infotrieve]
Sukhanov S, Song YH, Delafontaine P. Global analysis of differentially expressed genes in oxidized LDL-treated human aortic smooth muscle cells. Biochem Biophys Res Commun. 2003; 306: 443–449. [Order article via Infotrieve]
Li Y, Higashi Y, Itabe H, Song YH, Du J, Delafontaine P. Insulin-like growth factor-1 receptor activation inhibits oxidized LDL-induced cytochrome C release and apoptosis via the phosphatidylinositol 3 kinase/Akt signaling pathway. Arterioscler Thromb Vasc Biol. 2003; 23: 2178–2184.
Napoli C, de Nigris F, Palinski W. Multiple role of reactive oxygen species in the arterial wall. J Cell Biochem. 2001; 82: 674–682. [Order article via Infotrieve]
Nicholson AC, Han J, Febbraio M, Silversterin RL, Hajjar DP. Role of CD36, the macrophage class B scavenger receptor, in atherosclerosis. Ann N Y Acad Sci. 2001; 947: 224–228.
Jala VR, Haribabu B. Leukotrienes and atherosclerosis: new roles for old mediators. Trends Immunol. 2004; 25: 315–322. [Order article via Infotrieve]
Reddy MA, Kim YS, Lanting L, Natarajan R. Reduced growth factor responses in vascular smooth muscle cells derived from 12/15-lipoxygenase-deficient mice. Hypertension. 2003; 41: 1294–1300.
Cai H, Griendling KK, Harrison DG. The vascular NAD(P)H oxidases as therapeutic targets in cardiovascular diseases. Trends Pharmacol Sci. 2003; 24: 471–478. [Order article via Infotrieve]
Barry-Lane PA, Patterson C, van der Merwe M, Hu Z, Holland SM, Yeh ET, Runge MS. p47phox is required for atherosclerotic lesion progression in ApoE(-/-) mice. J Clin Invest. 2001; 108: 1513–1522.
Watanabe T, Pakala R, Katagiri T, Benedict CR. Oxidized low-density lipoproteins potentiate the mitogenic effect of 5-hydroxytryptamine on vascular smooth muscle cells. Jpn Heart J. 2002; 43: 35–42. [Order article via Infotrieve]
Suematsu M, Schmid-Schonbein GW, Chavez-Chavez RH, Yee TT, Tamatani T, Miyasaka M, Delano FA, Zweifach BW. In vivo visualization of oxidative changes in microvessels during neutrophil activation. Am J Physiol. 1993; 264: H881–H891. [Order article via Infotrieve]
Faraci FM, Didion SP. Vascular protection: superoxide dismutase isoforms in the vessel wall. Arterioscler Thromb Vasc Biol. 2004; 24: 1367–1373.
Schuppe-Koistinen I, Moldeus P, Bergman T, Cotgreave IA. S-thiolation of human endothelial cell glyceraldehyde-3-phosphate dehydrogenase after hydrogen peroxide treatment. Eur J Biochem. 1994; 221: 1033–1037.
Little C, O’Brien PJ. Mechanism of peroxide-inactivation of the sulphydryl enzyme glyceraldehyde-3-phosphate dehydrogenase. Eur J Biochem. 1969; 10: 533–538. [Order article via Infotrieve]
Stadtman ER. Metal ion-catalyzed oxidation of proteins: biochemical mechanism and biological consequences. Free Radic Biol Med. 1990; 9: 315–325. [Order article via Infotrieve]
Okura Y, Brink M, Itabe H, Scheidegger KJ, Kalangos A, Delafontaine P. Oxidized low-density lipoprotein is associated with apoptosis of vascular smooth muscle cells in human atherosclerotic plaques. Circulation. 2000; 102: 2680–2686.
Sukhanov STJ, Serrano C, Bowers C, Delafontaine P. Anti-atherogenic effects of growth hormone releasing peptide-2 in vitro: potential implications in prevention of atherosclerosis. American Heart Association Scientific Sessions, November 13–16, 2005, Dallas, Tex; 2005:651 Abstract.
Griendling KK, Minieri CA, Ollerenshaw JD, Alexander RW. Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells. Circ Res. 1994; 74: 1141–1148.
Mehrabian M, Allayee H, Wong J, Shi W, Wang XP, Shaposhnik Z, Funk CD, Lusis AJ. Identification of 5-lipoxygenase as a major gene contributing to atherosclerosis susceptibility in mice. Circ Res. 2002; 91: 120–126.
Huo Y, Zhao L, Hyman MC, Shashkin P, Harry BL, Burcin T, Forlow SB, Stark MA, Smith DF, Clarke S, Srinivasan S, Hedrick CC, Pratico D, Witztum JL, Nadler JL, Funk CD, Ley K. Critical role of macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 110: 2024–2031.
Santanam N, Aug N, Zhou M, Keshava C, Parthasarathy S. Overexpression of human catalase gene decreases oxidized lipid-induced cytotoxicity in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1999; 19: 1912–1917.
Muelleman RL, Grandstaff PM, Robinson WA. The use of pegorgotein in the treatment of frostbite. Wilderness Environ Med. 1997; 8: 17–19. [Order article via Infotrieve]
Schuppe-Koistinen I, Gerdes R, Moldeus P, Cotgreave IA. Studies on the reversibility of protein S-thiolation in human endothelial cells. Arch Biochem Biophys. 1994; 315: 226–234. [Order article via Infotrieve]
Tsuchiya Y, Yamaguchi M, Chikuma T, Hojo H. Degradation of glyceraldehyde-3-phosphate dehydrogenase triggered by 4-hydroxy-2-nonenal and 4-hydroxy-2-hexenal. Arch Biochem Biophys. 2005; 438: 217–222. [Order article via Infotrieve]
Divald A, Powell SR. Proteasome mediates removal of proteins oxidized during myocardial ischemia. Free Radic Biol Med. 2006; 40: 156–164. [Order article via Infotrieve]
Kiffin R, Christian C, Knecht E, Cuervo AM. Activation of chaperone-mediated autophagy during oxidative stress. Mol Biol Cell. 2004; 15: 4829–4840.
Molteni SN, Fassio A, Ciriolo MR, Filomeni G, Pasqualetto E, Fagioli C, Sitia R. Glutathione limits Ero1-dependent oxidation in the endoplasmic reticulum. J Biol Chem. 2004; 279: 32667–32673.
Levin M, Leppanen O, Evaldsson M, Wiklund O, Bondjers G, Bjornheden T. Mapping of ATP, glucose, glycogen, and lactate concentrations within the arterial wall. Arterioscler Thromb Vasc Biol. 2003; 23: 1801–1807.
Heinle H. Metabolite concentration gradients in the arterial wall of experimental atherosclerosis. Exp Mol Pathol. 1987; 46: 312–320. [Order article via Infotrieve]
您现在查看是摘要介绍页，详见ORG附件。