Proteasome-Dependent Degradation of Guanosine 5′-Triphosphate Cyclohydrolase I Causes Tetrahydrobiopterin Deficiency in Diabetes Mellitus
Background— Tetrahydrobiopterin (BH4) deficiency is reported to uncouple the enzymatic activity of endothelial nitric oxide synthase in diabetes mellitus. The mechanism by which diabetes actually leads to BH4 deficiency remains elusive. Here, we demonstrate that diabetes reduced BH4 by increasing 26S proteasome-dependent degradation of guanosine 5′-triphosphate cyclohydrolase I (GTPCH), a rate-limiting enzyme in the synthesis of BH4, in parallel with increased formation of both superoxide and peroxynitrite (ONOO−).
Methods and Results— Exposure of human umbilical vein endothelial cells to high glucose concentrations (30 mmol/L D-glucose) but not to high osmotic conditions (25 mmol/L L-glucose plus 5 mmol/L D-glucose) significantly lowered the levels of both GTPCH protein and BH4. In addition, high glucose increased both the 26S proteasome activity and the ubiquitination of GTPCH. Inhibition of the 26S proteasome with either MG132 or PR-11 prevented the high glucose–triggered reduction of GTPCH and BH4. Exposure of human umbilical vein endothelial cells to exogenous ONOO− increased proteasome activity and 3-nitrotyrosine in 26S proteasome. Furthermore, adenoviral overexpression of superoxide dismutase and inhibition of endothelial nitric oxide synthase with NG-nitro-l-arginine methyl ester significantly attenuated the high glucose–induced activation of 26S proteasome and the reduction of GTPCH. Finally, administration of MG132 or a superoxide dismutase mimetic, tempol, reversed the diabetes mellitus–induced reduction of GTPCH and BH4 and endothelial dysfunction in streptozotocin-induced diabetes mellitus.
Conclusions— We conclude that diabetes mellitus triggers BH4 deficiency by increasing proteasome-dependent degradation of GTPCH.
Received December 18, 2006; accepted June 15, 2007.
Many studies1–4 in patients and in vitro indicate that diabetes mellitus alters the metabolism and function of endothelium in ways that could lead to vascular injury. In diabetes mellitus, the function of endothelial nitric oxide (NO) synthase (eNOS) is altered such that this enzyme produces superoxide anion (O2–·) rather than nitric oxide (NO). This phenomenon is referred to as eNOS “uncoupling.”4 This transformation of eNOS from a protective enzyme to a contributor of oxidative stress has been observed in several in vitro models, in animal models of cardiovascular diseases, and in patients with cardiovascular risk factors (reviewed in Forstermann and Munzel1). Several studies5–7 have suggested that tetrahydrobiopterin (BH4), an essential cofactor for eNOS, might play a particularly important role in regulating NO and O2–· production by eNOS. In addition, recent studies8–10 have demonstrated that BH4 deficiency is responsible for eNOS uncoupling in diabetes mellitus, because these abnormalities were effectively prevented by coadministration of sepiapterin, a precursor for BH4, in diabetic animals8,9 and patients with type II diabetes mellitus10; however, the mechanism for BH4 reduction in diabetes remains poorly defined.
Clinical Perspective p 953
The BH4 pool in endothelial cells is variable, and a continuous supply is needed to maintain its basal levels. BH4 is synthesized from GTP de novo by the rate-limiting enzyme guanosine 5′-triphosphate cyclohydrolase I (GTPCH).11 As the first enzyme in the biosynthetic pathway of BH4, GTPCH is constitutively expressed in endothelial cells. This enzyme is critical to the maintenance of BH4 levels because its inhibition leads to a rapid decrease of BH4. GTPCH enzyme activity is regulated by several mechanisms, which vary among different cell types and include protein expression, posttranslational modifications,12 and association with the GTPCH feedback regulatory protein (GFRP).13 In the insulin-resistance rat model, GTPCH activity in the aorta was significantly lower than that of control rats.14 In GTPCH transgenic mice, which showed a selective augmentation of endothelial BH4 levels, diabetes-enhanced O2–· release was reduced significantly, whereas NO-mediated vasodilatation was preserved.6 More recent studies in the deoxycorticosterone acetate–salt hypertensive mouse also demonstrated that decreased BH4 levels were related to reduced GTPCH activity15 and that supplementation of BH4 or increased BH4 synthesis by adenoviral gene delivery of GTPCH restored BH4 levels and normalized eNOS function15; however, the exact mechanisms regulating GTPCH in the endothelium remain poorly understood.
Recent evidence indicates that the endothelial dysfunction associated with diabetes mellitus (or in the case of in vitro studies, in response to high glucose concentrations) results from the local formation of oxidants and free radicals within and in the vicinity of the vascular endothelium.2,16 Evidence suggests that oxidative stress is sufficient to produce eNOS uncoupling. In deoxycorticosterone acetate–salt hypertension, p47phox knockout mice were relatively protected from eNOS uncoupling,15 which suggests that NADPH oxidase–mediated O2–· production is an important contributor to eNOS uncoupling. Other interventions, such as angiotensin II receptor blockade17 or inhibition of protein kinase C,18 that reduce NAD(P)H oxidase activation in vascular disease states have also been shown to reduce eNOS uncoupling and enhance NO bioavailability. These observations have contributed to a paradigm whereby O2–· in the endothelium (principally from NADPH oxidase) is thought to generate ONOO− (by reacting with NO). Ultimately, this generation leads to eNOS uncoupling and, thus, perpetuates a cycle of vascular oxidative stress. Because oxidant stress is the common mechanism in the pathogenesis of diabetic vascular disease, we hypothesized that O2–· and ONOO−, on reaction with NO, would uncouple eNOS by reducing the levels of BH4 as a result of GTPCH reduction. Here, we demonstrate that hyperglycemia via NAD(P)H oxidase–derived O2–· and ONOO− activated the ubiquitin-proteasome pathway and resulted in accelerated GTPCH degradation and consequent BH4 deficiency.
Polyclonal antibodies against human GTPCH were kindly provided by Dr Gabriele Werner-Felmayer (Innsbruck Medical University, Innsbruck, Austria). Monoclonal anti-mouse GTPCH antibody was purchased from Ascenion GmbH (Munich, Germany). Antibody against 3-nitrotyrosine (3-NT) was obtained from Upstate Biotechnology, Inc (Lake Placid, NY). Antibody against ubiquitin was obtained from Santa Cruz Biotechnology, Inc (Santa Cruz, Calif). Mouse anti-26S proteasome antibody was obtained from Abcam, Inc (Cambridge, Mass). The proteasome inhibitors MG132 and PR-11 were purchased from BioMol (Plymouth Meeting, Pa). Fluorogenic proteasome substrates were purchased from Calbiochem (San Diego, Calif). Protein-A sepharose CL-4B was purchased from Amersham Biosciences (Piscataway, NJ). Dihydroethidium (DHE) was purchased from Molecular Probes (Eugene, Ore). BH4 was obtained from Cayman Chemical Company (Ann Arbor, Mich). Tetrahydro-l-biopterin dihydrochloride was purchased from Calbiochem. RNA UltraSense One-Step qRT-PCR System was obtained from Invitrogen (Carlsbad, Calif). The other chemicals, unless otherwise noted, were obtained from Fisher Scientific at the highest grade of quality (Pittsburgh, Pa).
Ten-week-old C57BL/6J mice were obtained from the Jackson Laboratory (Bar Harbor, Me). Mice were housed in temperature-controlled cages with a 12-hour light/dark cycle and given free access to water and food. The animal protocol was reviewed and approved by the Institutional Animal Care and Use Committee at the University of Oklahoma Health Sciences Center.
Human umbilical vein endothelial cells (HUVECs) and cell culture media were obtained from Cascade Biologics (Walkersville, Md). Confluent cells grown in 6-well plates or 100-mm dishes were used between passages 2 and 4. Before all treatments, endothelial cells were grown in low serum for 2 hours.
Confluent HUVECs were infected with either adenovirus encoding manganese superoxide dismutase (Mn-SOD) or dominant negative p67phox (p67phox-DN), as described previously.19 Adenovirus encoding green fluorescent protein was used as a control.
Detection of Reactive Oxygen Species
Intracellular O2–· was assessed by the DHE fluorescence/high-performance liquid chromatography assay20 with minor modifications. Briefly, confluent HUVECs were incubated with DHE (0.5 μmol/L) for 30 minutes before cell harvest, and then the cells were subjected to methanol extraction. High-performance liquid chromatography was used to separate and quantify oxyethidium (a product of DHE and O2–·) and ethidium (a product of DHE auto-oxidation) with a C-18 column (mobile phase: gradient of acetonitrile and 0.1% trifluoroacetic acid). O2–· production was determined by the conversion of DHE into oxyethidium.
Detection of BH4 and Total Biopterins
The levels of BH4 and total biopterins were determined as described previously21 with some modification. Briefly, HUVECs (3 to 5×106 cells) were washed and suspended in distilled water containing 5 mmol/L dithioerythrol, centrifuged at 12 000g at 4°C for 10 minutes, and then subjected to oxidation in acid or base. To a 100 μL aliquot of supernatant, 20 μL of 0.5 mol/L HCl and 0.05 mol/L iodine was added for acidic oxidation, and 20 μL of 0.5 mol/L NaOH plus 0.05 mol/L iodine was added for basic oxidation. After incubation for 1 hour in the dark at room temperature, 20 μL of HCl was added to the basic oxidation only. All mixtures received 20 μL of 0.1 mol/L ascorbic acid for the reduction of excess iodine. Samples were then centrifuged for 10 minutes at 12 000g at 4°C. Biopterin concentrations were determined by high-performance liquid chromatography with a PR-C18 column. Elution was at a rate of 1.0 mL of 50 mmol/L potassium phosphate buffer per minute (pH 3.0). Fluorescence was detected with excitation at 350 nm and emission at 440 nm. BH4 concentrations were calculated as the difference in results from oxidation in acid and base.
26S Proteasome Activity Assay
The 26S proteasome function was measured as described previously.22 Briefly, cells were washed with PBS and then with buffer I (50 mmol/L Tris, pH 7.4, 2 mmol/L DTT, 5 mmol/L MgCl2, 2 mmol/L ATP). The cells were then made into pellets by centrifugation. Homogenization buffer (50 mmol/L Tris, pH 7.4, 1 mmol/L DTT, 5 mmol/L MgCl2, 2 mmol/L ATP, 250 mmol/L sucrose) was added, and cells were subjected to a vortex for 1 minute. Cell debris was removed by centrifugation at 1000g for 5 minutes followed by 10 000g for 20 minutes. Protein concentration was determined by the BCA (bicinchoninic acid) protocol (Pierce, Rockford, Ill). Protein (100 μg) from each sample was diluted with buffer I to a final volume of 1000 μL. The fluorogenic proteasome substrate Suc-LLVY-7-amido-4-methylcoumarin (chymotrypsin-like, Sigma, St. Louis, Mo) was added at a final concentration of 80 μmol/L in 1% DMSO. Cleavage activity was monitored continuously by detection of free 7-amido-4-methylcoumarin with a fluorescence plate reader (Gemini, Molecular Devices, Sunnyvale, Calif) at 380/460 nm at 37°C.
Assay of GTPCH Activity
GTPCH activity in HUVECs or in aorta preparations was measured as reported previously.23 This assay was based on the quantification of D-erythro-neopterin by high-performance liquid chromatography after the conversion of enzymatically formed D-erythro-7,8-dihydroneopterin triphosphate into D-erythro-neopterin by sequential iodine oxidation and dephosphorylation.
Reverse-Transcription Polymerase Chain Reaction for GTPCH
RNA was isolated from the treated HUVECs with the total RNA isolation protocol for the RNeasy Mini Kit (Qiagen Inc, Valencia, Calif). Reverse-transcription polymerase chain reaction was performed according to the manufacturer’s protocol (RNA UltraSense One-Step qRT-PCR System, Invitrogen). Polymerase chain reaction primer pairs were as described previously.24
Streptozotocin-Induced Diabetes Mellitus in Mice
A low-dose streptozotocin (STZ) induction regimen was used to induce pancreatic islet cell destruction and persistent hyperglycemia as previously described by the Animal Models of Diabetic Complications Consortium (http://www.amdcc.org). Hyperglycemia was defined as a random blood glucose level of >450 mg/dL for >2 weeks after injection. One additional group of STZ mice received intraperitoneal MG132 injections (5 mg/kg body weight) for 2 days as described previously.25 Another group of STZ mice received tempol (Sigma; 1 mmol/L) in their drinking water for an additional 2 weeks. Aortas were all harvested 3 weeks after STZ injection. Mouse aortas were removed and prepared as reported previously.26
Enrichment of Ubiquitinated Proteins
Ubiquitinated proteins were enriched with affinity beads (ubiquitinated protein enrichment kit, catalog No. 662200, Calbiochem). These beads were conjugated with the ubiquitin-associated sequence, which is known to form domains that bind ubiquitinated cellular proteins. Enrichment of ubiquitinated proteins was performed according to the manufacturer’s instructions.
Organ Chamber Experiment With Mouse Aorta
Aortas were isolated from mice treated with vehicle, STZ, and STZ/MG132, cut into 3-mm rings, and precontracted with 30 nmol/L of U46619 in organ chambers (PowerLab, ADInstruments, Colorado Springs, Colo). Endothelium-dependent and -independent vasodilator responses were determined in the presence of acetylcholine (0.01 to 100 μmol/L) and sodium nitroprusside (0.0001 to 1 μmol/L), respectively.
Statistical comparisons of vasodilation (organ chamber experiments) were performed with 2-way ANOVA, and intergroup differences were tested with Bonferroni inequality. All other results were analyzed with 1-way ANOVA, except for those obtained from the time-course studies, which were analyzed with repeated-measures ANOVA. Values are expressed as mean±SEM for all assays. Significance was accepted at P<0.05.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
High Glucose Decreases Both Total Biopterins and BH4 in HUVECs
To investigate whether hyperglycemia alters the levels of both total biopterins and BH4, confluent HUVECs were exposed to normal glucose (NG; 5 mmol/L D-glucose), high glucose (HG; 30 mmol/L D-glucose), or high osmotic L-glucose (25 mmol/L L-glucose plus 5 mmol/L D-glucose) concentrations for 2 hours. After incubation, HUVECs were assayed for total biopterin and BH4 levels. Compared with NG, exposure of HUVECs to HG but not to high osmotic L-glucose concentrations significantly lowered the levels of both total biopterins (BH4 plus dihydrobiopterin [BH2] plus other biopterins; Figure 1A, open bars) and BH4 (−25.0±5.7%, n=6, P<0.01; Figure 1A, solid bars). Furthermore, the HG-induced BH4 reduction was time dependent (Figure 1B, solid bars). HG did not alter the ratio of BH4/oxidized biopterin (BH2 plus other biopterins; NG 2.6±0.5 versus HG 2.5±0.7, P>0.5), which suggests that the HG-induced BH4 reduction was mainly due to a decrease in the synthesis of biopterins, probably via a reduction in GTPCH activity.
HG Reduces GTPCH Levels and Activity in HUVECs
GTPCH is the rate-limiting enzyme for the production of biopterins, including BH4. Previous studies27,28 have demonstrated that diabetes mellitus suppresses the levels of GTPCH. We next examined whether HG precipitated a reduction in GTPCH activity. Determination of the rate of neopterin conversion from GTP showed that HG but not NG significantly inhibited the activity of GTPCH in HUVECs in a time-dependent manner (Figure 1B). In addition, HG but not high osmotic L-glucose concentrations significantly lowered the protein levels of GTPCH (Figure 1C) in HUVECs. The reduction in GTPCH was seen as early as 1 hour after HG exposure (Figure 1D). Two hours after exposure, HG significantly reduced the level of GTPCH (−65±3%, n=6, P<0.01; Figure 1D). In contrast, HG did not alter the protein level of GFRP, a negative regulator of GTPCH (data not shown), or its association with GTPCH (Figure 1E). Furthermore, HG did not alter GTPCH gene expression, as determined by the levels of GTPCH mRNA expression in HUVECs (Figure 1F). Taken together, these results suggest that HG suppresses GTPCH activity via a decrease in GTPCH protein levels.
Proteasome Inhibition Abolishes HG-Induced Reduction of Both GTPCH and BH4 in HUVECs
Because HG lowered GTPCH protein levels without altering its expression, we next determined whether HG suppressed GTPCH by increasing GTPCH degradation via the 26S proteasome. To this end, MG132 and PR-11, 2 potent but structurally distinct 26S proteasome inhibitors, were incubated with HUVECs for 1 hour before HG exposure. As depicted in Figure 2A, neither MG132 (0.5 μmol/L) nor PR-11 (0.5 μmol/L) altered basal levels of GTPCH in HUVECs; however, administration of either of these inhibitors abrogated the HG-induced GTPCH reduction in HUVECs (Figure 2A). Furthermore, both MG132 and PR-11 abolished the HG-induced inhibition of GTPCH activity (Figure 2B). Importantly, although neither inhibitor altered the basal BH4 levels, MG132 (Figure 2C) and PR-11 (data not shown) prevented the HG-induced BH4 reduction.
HG Increases GTPCH Ubiquitination and 26S Proteasome Activity
Increasing evidence suggests that most 26S proteasome-dependent protein degradation is mediated by protein ubiquitination, a process by which ubiquitin is covalently attached to the ε-amino group of lysine residues in a substrate protein.29 Protein ubiquitination can be monitored via Western blot with a specific antibody against ubiquitin.30 Thus, we examined whether HG increases the ubiquitination of GTPCH in HUVECs exposed to HG. Cell lysates were subjected to ubiquitin enrichment, and subsequently, the lysates were subjected to Western blot analysis with an anti-GTPCH antibody. As shown in Figure 3, exposure of HUVECs to HG for 1 hour, which did not alter the total GTPCH, significantly increased the ubiquitination of GTPCH (Figure 3A). Prolonged incubation (>1 hour) caused a decline in the total amount of GTPCH and in the ubiquitinated GTPCH pool in HUVECs. This finding suggests GTPCH is degraded by a ubiquitin-dependent mechanism.
Treatment with MG132 alone increased the staining of ubiquitinated GTPCH (Figure 3A), which indicates the involvement of the 26S proteasome in maintaining GTPCH protein turnover in resting HUVECs. Moreover, MG132 prevented the HG-induced decrease in GTPCH protein levels but did not alter the levels of ubiquitinated GTPCH in HUVECs (Figure 3A).
We next examined whether HG increases the 26S proteasome activity in HUVECs. Total 26S proteasome activity in HUVECs was assayed with specific fluorescent proteasome substrates. As depicted in Figure 3B, exposure of HUVECs to HG but not NG significantly increased 26S proteasome activity by 3-fold. Taken together, these results indicate that HG increased the ubiquitination and subsequent degradation of GTPCH in HUVECs via increased activity of the 26S proteasome.
HG Increases Both O2–· and ONOO− in HUVECs
Long-term exposure of endothelial cells to HG triggers endothelial dysfunction by releasing both enhanced O2–· and ONOO−.31,32 We next determined whether short-term exposure of HUVECs to HG increased O2–· and ONOO−. Exposure of HUVECs to HG for 2 hours significantly increased the formation of oxyethidium, the specific product of the reaction of DHE with O2–· (Figure 4A), which indicates that HG increased O2–· in HUVECs.
Endothelial cells simultaneously release both O2–· and NO, which combine into the potent oxidant ONOO−.3 3-NT is considered the “footprint” of ONOO− in cultured cells. To assay whether HG increased the formation of 3-NT–positive proteins in HUVECs, equal amounts of cell lysate were subjected to Western blot analysis with a specific anti-3-NT antibody. HG but not NG significantly increased the levels of 3-NT–positive proteins 2 hours after HG exposure (Figure 4A). Thus, acute hyperglycemia increases both O2–· and ONOO− in HUVECs.
HG-Induced GTPCH Reduction Is ONOO− Dependent
We examined whether HG triggered the reduction of GTPCH in HUVECs via ONOO−. HUVECs were treated with either chemically synthesized ONOO− or ONOO− donor SIN-1 (linsidomine chlorhydrate) for 30 minutes. As shown in Figure 4B, exposure of HUVECs to either synthetic ONOO− (50 μmol/L) or SIN-1 (100 μmol/L) significantly decreased the GTPCH level similar to HG treatment.
We next determined whether uric acid, which is a potent scavenger of ONOO−, prevented the HG-induced reduction in GTPCH protein levels and activity in HUVECs. Uric acid (100 μmol/L) alone did not alter either basal GTPCH expression (data not shown) or GTPCH activity; however, treatment with this chemical abolished the HG-induced reduction of GTPCH protein levels (Figure 4B) and activity (Figure 4C).
ONOO− is formed during simultaneous generation of O2–· and NO.33 Inhibition of either O2–· or NO production abolishes the formation of ONOO−. Adenoviral overexpression of SOD, an O2–· scavenger (Figure 4E), or administration of NG-nitro-l-arginine methyl ester (L-NAME; 1 mmol/L), an NO synthase inhibitor (Figure 4D), abolished the HG-induced GTPCH reduction, which suggests that GTPCH reduction is in fact dependent on ONOO−.
We next determined the source of ONOO− in HG-induced GTPCH reduction. Both pharmacological inhibition of NAD(P)H oxidase assembly with apocynin (10 μmol/L), which inhibits NADPH oxidase–derived O2–· (Figure 4D), and genetic inhibition of NAD(P)H oxidase via overexpression of the dominant negative version (p47phox-DN) of NAD(P)H oxidase (Figure 4E) abolished HG-induced GTPCH reduction. Thus, the NAD(P)H oxidases may be the source of O2–· and ONOO− in HUVECs exposed to HG.
ONOO− Activates 26S Proteasome in HUVECs
To determine whether ONOO− was responsible for the HG-enhanced 26S proteasome activity and consequent GTPCH reduction, the activity of the 26S proteasome was assayed in HUVECs exposed to SIN-1, an exogenous ONOO− donor. SIN-1 alone significantly increased 26S proteasome activity by 3-fold (Figure 5A). In addition, uric acid (100 μmol/L), a potent ONOO− scavenger, markedly prevented SIN-1–enhanced 26S proteasome activation (Figure 5A), whereas uric acid alone did not alter the basal activity of the proteasome (Figure 5B), which suggests that ONOO− mimics the effects of HG on the 26S proteasome.
To further examine whether ONOO− mediated the HG-induced activation of the 26S proteasome in intact cells, HUVECs were preincubated with pharmacological reagents to inhibit the ONOO−-forming components O2–· or NO before exposure to HG. Indeed, the presence of apocynin (10 μmol/L), which is an inhibitor of NADPH oxidase–derived O2–·, and L-NAME (1 mmol/L), which is an inhibitor of NO synthase–derived NO, abrogated the HG-induced increase in 26S proteasome activity (Figure 5B).
Tyrosine Nitration of the 26S Proteasome in HUVECs Exposed to HG
To study the potential mechanism underlying the HG-induced ONOO−-mediated proteasome activity, cell lysates from HUVECs treated with HG for 0.5 hour were immunoprecipitated with an anti-26S proteasome antibody and assayed for 3-NT–positive proteins by Western blotting with an anti-3-NT antibody. Compared with cells exposed to NG, HUVECs exposed to HG exhibited a significant increase in 3-NT–positive proteins, which indicates increased nitration of 26S proteasome (Figure 5C). Furthermore, treatment with uric acid, a potent scavenger of ONOO−, significantly attenuated HG-induced nitration of 26S proteasome in HUVECs. This result indicates that HG activates the 26S proteasome via ONOO−-dependent nitration.
Proteasome Inhibition With MG132 or a SOD Mimetic, Tempol, Prevents Diabetes Mellitus–Induced Reduction of Both GTPCH and BH4 in STZ-Induced Diabetic Mice
We further investigated the reduction of GTPCH and BH4 during diabetes mellitus in vivo. Diabetes mellitus was induced in mice by STZ injection, and mice were subsequently treated with MG132 and a mimetic SOD, tempol. Injection of STZ increased serum glucose to an average of 450±20 mg/dL (data not shown), whereas the control group receiving only vehicle exhibited glucose levels of 180±15 mg/dL (n=5, P<0.01; data not shown). Administration of MG132 for 2 days or treatment with tempol in the drinking water for 2 weeks did not alter the blood glucose levels of control mice (n=5, P>0.5) or STZ-induced diabetic mice (n=5, P>0.5). No significant difference in body weight was observed between these groups during the experimental period (data not shown).
We first determined whether diabetes increased ONOO− in vivo. 3-NT, a marker of ONOO−, was measured in homogenates of aortas prepared from mice from all experimental groups. STZ treatment, but not treatment with vehicle, MG132, or tempol alone, increased the levels of 3-NT–positive proteins as detected in Western blots with an anti-3-NT antibody (Figure 6A). As expected, tempol treatment but not MG132 treatment attenuated the diabetes mellitus–enhanced 3-NT in STZ-injected mice (Figure 6A). Consistently, STZ, but not vehicle, MG132, or tempol alone, increased 26S proteasome activity by ≈2-fold (Figure 6B). Similarly, tempol, via scavenging of O2–·, consequently reduced ONOO− and inhibited 26S proteasome activation in the diabetes mellitus model.
STZ-induced hyperglycemia actually decreased GTPCH levels (Figure 6C). This effect was reversed by treatment with either MG132 or tempol (Figure 6C). Furthermore, STZ-induced diabetes mellitus decreased GTPCH activity and BH4 levels (Figure 6D). In contrast, either MG132 or tempol (Figure 6D) treatment of STZ-treated mice restored both GTPCH activity and BH4 levels in vivo. Taken together, these data demonstrate that O2–· and ONOO− caused by diabetes mellitus activated the ubiquitin-proteasome pathway and resulted in accelerated GTPCH degradation and consequent BH4 deficiency in vivo.
Proteasome Inhibition With MG132 Reverses STZ-Induced Endothelial Dysfunction
We next investigated whether MG132 reversed diabetes mellitus–induced endothelial dysfunction in vivo. Endothelium-dependent relaxation was assayed in isolated mouse aortas treated with either vehicle or STZ, with or without MG132. As shown in Figure 6E, diabetes mellitus impaired acetylcholine-induced endothelium-dependent relaxation. Furthermore, administration of MG132 in diabetic mice abolished diabetes mellitus–impaired endothelium-dependent relaxation but had no effect on NO donor–triggered endothelium-independent relaxation (Figure 6F). Because eNOS is the major source of NO that maintains endothelial function, these results indicate that MG132 might restore eNOS function by inhibiting the proteasome-dependent GTPCH reduction caused by diabetes mellitus.
In the present study, we have demonstrated that diabetes mellitus reduces levels of BH4, an essential cofactor for eNOS, via 26S proteasome–mediated degradation of GTPCH, which is the rate-limiting enzyme of BH4 synthesis. Additionally, we found that HG significantly increased O2–· and 3-NT–positive proteins and that adenoviral overexpression of SOD or inhibition of eNOS with L-NAME significantly attenuated HG-induced 26S proteasome activation and GTPCH reduction. Thus, it is likely that the HG-induced proteasome activation and GTPCH reduction are mediated by endogenous ONOO−. Moreover, administration of either the proteasome inhibitor MG132 or the SOD mimetic tempol reversed the reduction of both GTPCH and BH4 in STZ-induced diabetes mellitus in mice. Finally, MG132 abolished diabetes mellitus–induced endothelial dysfunction in vivo. Therefore, BH4 deficiency in diabetes mellitus is due to a reduction in GTPCH, an enzyme critical to BH4 synthesis, via a process that is peroxynitrite mediated and proteasome dependent. These results may have uncovered a novel mechanism underlying BH4 deficiency in diabetic vascular diseases.
The majority of protein degradation in mammalian cells is catalyzed by a multistep pathway that requires ATP hydrolysis, the protein cofactor ubiquitin, and the 26S proteasome. This process offers varied rates of degradation of different cellular proteins that can be regulated precisely.29,34 Ubiquitin and ubiquitin-like modifications are increasingly recognized as key regulatory events in health and disease. This ATP-dependent ubiquitin-proteasome system has evolved as the premier cellular proteolytic machinery. As a result, dysregulation of this system by several different mechanisms leads to inappropriate degradation of specific proteins and pathological consequences.34 Although aberrations in the ubiquitin-proteasome pathway have been implicated in certain malignancies and neurodegenerative disorders, the pathological implications of this system in the pathogenesis of diabetes mellitus and its complications have not been elucidated completely. The present study has provided the first evidence that hyperglycemia activates the 26S proteasome via ONOO−-dependent mechanisms and leads to GTPCH degradation. Major evidence in support of this notion includes the following: (1) the structurally distinct yet specific proteasome inhibitors MG132 and PR-11 efficiently blocked the HG-induced degradation of GTPCH and the consequent BH4 deficiency; (2) HG induced downregulation of GTPCH and BH4 without changing the GTPCH mRNA levels; (3) HG activated proteasome activity by 3-fold and induced proteasome nitration by ONOO−; (4) HG essentially enhanced GTPCH ubiquitination, which made GTPCH a good substrate for proteasome degradation; (5) inhibition by an ONOO− scavenger, which abolishes HG-induced GTPCH degradation and BH4 downregulation, inhibited the HG-enhanced proteasome activity; and (6) STZ-induced hyperglycemia decreased GTPCH and BH4, and this effect was reversed by MG132 treatment without alteration of blood glucose levels. In fact, recent evidence in a rat model of diabetes mellitus suggests that reduced BH4 results from the decreased expression and activity of GTPCH.27,35 Consistent with these results, growing evidence has indicated that GTPCH gene transfer increases BH4 bioavailability and restores endothelial function.5–7,14,36 As a result, a crucial role for GTPCH level and/or activity is necessary to maintain BH4 bioavailability and endothelial function both in physiological and pathological conditions. Acceleration of GTPCH degradation by the ubiquitin-proteasome pathway as described here may have important implications for the development of vascular endothelial dysfunction and the evolution of atherosclerotic plaques in patients with diabetes mellitus. Indeed, a previous postmortem study37 demonstrated enhanced ubiquitin expression in unstable coronary plaques. Furthermore, higher expression of ubiquitin, along with increased activation of proteasome, was found in human plaque macrophages obtained from the asymptomatic carotid lesions of patients with type 2 diabetes mellitus compared with lesions from nondiabetic patients.38
Although evidence39 suggests that proteasomes, especially 26S proteasome, are sensitive to oxidants such as H2O2, hypochlorite, and ONOO−, the mechanism by which oxidative stress alters the functions of proteasome is not fully clarified. In the present study, we found that ONOO− activated 26S proteasome and increased 3-NT staining in the proteasome. Importantly, tempol, which prevented ONOO− generation from HG, not only inhibited diabetes mellitus–enhanced proteasome activity but also restored levels of GTPCH activity, protein levels, and BH4 in STZ-treated mice. In addition, either scavenging ONOO− formation with uric acid or by blocking the generation of either O2–· or NO prevented the HG-induced GTPCH reduction. These results suggest that oxidants such as ONOO− play essential roles in regulating the function of proteasomes. Consistent with these findings, oxidative stress has been shown to stimulate the ubiquitin pathway in macrophages by inducing the expression of components of its enzymatic machinery, such as ubiquitin-binding proteins.40 In a recently published randomized, controlled trial,38 plaques from diabetic patients were shown to have more ubiquitin and proteasome activity, along with more markers of oxidative stress (3-NT and O2–· production), than plaques from control patients. Interestingly, rosiglitazone, a drug that suppressed both 3-NT and O2–· production in atherosclerotic plaques, had less ubiquitin and proteasome activity, which suggests that oxidant stress might contribute to the activation of proteasomes in atherosclerotic plaques in diabetes mellitus. Taken together, these data strongly support the notion that oxidants such as ONOO− contribute to proteasome activation in diabetes mellitus in vivo.
In diabetes mellitus, selective or enhanced degradation of targeted proteins, such as muscle proteins,41 mutated water channel aquaporin-2,42 insulin receptor substrates 1 and 2, and even insulin,43 occurs through an altered ubiquitin-proteasome system. Furthermore, exposure of isolated proteasome to oxidants, such as ONOO−, leads to alteration of proteasome activation,44 whereas exposure of certain cellular proteins to ONOO− leads to increased oxidation followed by preferential degradation by the proteasomal system.45 The mechanism of preferential degradation of GTPCH in diabetes mellitus, however, remains unknown. Because inhibition of ONOO− prevented GTPCH reduction in the present model of diabetes mellitus, we speculate that GTPCH might react with ONOO− and become modified, resulting in selective recognition and degradation by activated proteasome. This conclusion is supported by the fact that the activity of GTPCH decreased in parallel with BH4 reduction (Figure 1B) and that proteasome inhibitors (MG132 and PR-11) restored the GTPCH protein, GTPCH activity, and BH4 (Figure 2). Furthermore, MG132 prevented diabetes mellitus–induced vascular endothelium dysfunction in vivo (Figure 6). Interestingly, the marked reduction in GTPCH protein in HG is accompanied by a much more modest reduction in GTPCH activity and BH4 levels. The reason for this discordance remains unknown and warrants further investigation.
BH4 deficiency has been reported to play a causal role in the development of endothelial dysfunction in many vascular diseases in which “eNOS uncoupling” is present (reviewed in Moens and Kass46). Recent evidence supports potential cardiovascular benefits as a result of BH4 replacement for the treatment of vascular conditions that exhibit BH4 depletion, such as hypertension, ischemia-reperfusion injury, and cardiac hypertrophy with chamber remodeling (reviewed in Moens and Kass46). Recent studies in vitro showed that BH4 can be rapidly oxidized by reactive oxygen species such as ONOO−. Oxidation of BH4 is enhanced and vascular tissue levels of 7,8-dihydrobiopterin increase in several disease models, including DOCA-salt hypertensive mice and STZ-induced diabetes mellitus.4,6,7,15 In contrast to the previous findings, we did not find increased BH4 oxidation, even though we used the same detection method as other groups. This discrepancy is not likely to be explained by the various sensitivities of the detection method used, because we were able to measure the increased oxidation of BH4 into 7,8-dihydrobiopterin when exposed to authentic ONOO− (Xu et al, unpublished data) and in bovine aortic endothelial cells after being exposed to HG for 1 to 3 days (online Data Supplement Figure I). The discrepancy may be explained by the high sensitivity of GTPCH or 26S proteasome to the low influx of ONOO− generated in the early stage of diabetes mellitus. ONOO− might reduce the levels of BH4 by 2 interdependent mechanisms in diabetes mellitus: the inhibition of BH4 synthesis in the early phase, followed by increased BH4 oxidation in the late phase. Modulation of BH4 status in human vascular disease clearly represents a promising target for therapeutic interventions aimed at prevention of atherosclerotic disease. Because inhibition of GTPCH reduction by proteasome inhibitors may increase levels of BH4, proteasome inhibitors might be useful tools for the maintenance of BH4 levels, which consequently prevent eNOS uncoupling in cardiovascular diseases, including diabetes mellitus.
In summary, we have demonstrated that diabetic hyperglycemia activates the 26S proteasome via ONOO− and results in the ubiquitination and degradation of GTPCH, the limiting enzyme for BH4 synthesis, and subsequent impaired endothelium-dependent vessel relaxation (Figure 7). These impairments can be reversed by treatment with a proteasome inhibitor. These results are particularly relevant to hyperglycemic conditions and endothelial dysfunction, both of which are prevalent in type 1 and type 2 diabetes mellitus.
The authors acknowledge Dr Gabriele Werner-Felmayer for providing the specific anti-GTPCH antibody.
Sources of Funding
This work was supported by National Institutes of Health grants (HL079584, HL080499, and HL074399), a research award from the American Diabetes Association, a research award from the Juvenile Diabetes Research Foundation, a grant from the Oklahoma Center for the Advancement of Science and Technology (OCAST), and the Paul H. Doris Eaton Travis Chair Funds in Endocrinology of the University of Oklahoma Health Sciences Center (all to Dr Zou).
Forstermann U, Munzel T. Endothelial nitric oxide synthase in vascular disease: from marvel to menace. Circulation. 2006; 113: 1708–1714.
Brownlee M. The pathobiology of diabetic complications: a unifying mechanism. Diabetes. 2005; 54: 1615–1625.
Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Alp NJ, McAteer MA, Khoo J, Choudhury RP, Channon KM. Increased endothelial tetrahydrobiopterin synthesis by targeted transgenic GTP-cyclohydrolase I overexpression reduces endothelial dysfunction and atherosclerosis in ApoE-knockout mice. Arterioscler Thromb Vasc Biol. 2004; 24: 445–450.
Alp NJ, Mussa S, Khoo J, Cai S, Guzik T, Jefferson A, Goh N, Rockett KA, Channon KM. Tetrahydrobiopterin-dependent preservation of nitric oxide-mediated endothelial function in diabetes by targeted transgenic GTP-cyclohydrolase I overexpression. J Clin Invest. 2003; 112: 725–735.
Elrod JW, Duranski MR, Langston W, Greer JJ, Tao L, Dugas TR, Kevil CG, Champion HC, Lefer DJ. eNOS gene therapy exacerbates hepatic ischemia-reperfusion injury in diabetes: a role for eNOS uncoupling. Circ Res. 2006; 99: 78–85.
Nie H, Wu JL, Zhang M, Xu J, Zou MH. Endothelial nitric oxide synthase-dependent tyrosine nitration of prostacyclin synthase in diabetes in vivo. Diabetes. 2006; 55: 3133–3141.
Katusic ZS, d’Uscio LV. Tetrahydrobiopterin. mediator of endothelial protection. Arterioscler Thromb Vasc Biol. 2004; 24: 397–398.
Alp NJ, Channon KM. Regulation of endothelial nitric oxide synthase by tetrahydrobiopterin in vascular disease. Arterioscler Thromb Vasc Biol. 2004; 24: 413–420.
Xie L, Smith JA, Gross SS. GTP cyclohydrolase I inhibition by the prototypic inhibitor 2, 4-diamino-6-hydroxypyrimidine: mechanisms and unanticipated role of GTP cyclohydrolase I feedback regulatory protein. J Biol Chem. 1998; 273: 21091–21098.
Thai H, Wollmuth J, Goldman S, Gaballa M. Angiotensin subtype 1 receptor (AT1) blockade improves vasorelaxation in heart failure by up-regulation of endothelial nitric-oxide synthase via activation of the AT2 receptor. J Pharmacol Exp Ther. 2003; 307: 1171–1178.
Li H, Witte K, August M, Brausch I, Godtel-Armbrust U, Habermeier A, Closs EI, Oelze M, Munzel T, Forstermann U. Reversal of endothelial nitric oxide synthase uncoupling and up-regulation of endothelial nitric oxide synthase expression lowers blood pressure in hypertensive rats. J Am Coll Cardiol. 2006; 47: 2536–2544.
Bachschmid M, Thurau S, Zou MH, Ullrich V. Endothelial cell activation by endotoxin involves superoxide/NO-mediated nitration of prostacyclin synthase and thromboxane receptor stimulation. FASEB J. 2003; 17: 914–916.
Fink B, Laude K, McCann L, Doughan A, Harrison DG, Dikalov S. Detection of intracellular superoxide formation in endothelial cells and intact tissues using dihydroethidium and an HPLC-based assay. Am J Physiol Cell Physiol. 2004; 287: C895–C902.
Nar H, Huber R, Auerbach G, Fischer M, Hosl C, Ritz H, Bracher A, Meining W, Eberhardt S, Bacher A. Active site topology and reaction mechanism of GTP cyclohydrolase I. Proc Natl Acad Sci U S A. 1995; 92: 12120–12125.
Xu J, Xie Z, Reece R, Pimental D, Zou MH. Uncoupling of endothelial nitric oxidase synthase by hypochlorous acid: role of NAD(P)H oxidase-derived superoxide and peroxynitrite. Arterioscler Thromb Vasc Biol. 2006; 26: 2688–2695.
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.
Lecker SH, Goldberg AL, Mitch WE. Protein degradation by the ubiquitin-proteasome pathway in normal and disease states. J Am Soc Nephrol. 2006; 17: 1807–1819.
Meininger CJ, Cai S, Parker JL, Channon KM, Kelly KA, Becker EJ, Wood MK, Wade LA, Wu G. GTP cyclohydrolase I gene transfer reverses tetrahydrobiopterin deficiency and increases nitric oxide synthesis in endothelial cells and isolated vessels from diabetic rats. FASEB J. 2004; 18: 1900–1902.
Ihlemann N, Rask-Madsen C, Perner A, Dominguez H, Hermann T, Kober L, Torp-Pedersen C. Tetrahydrobiopterin restores endothelial dysfunction induced by an oral glucose challenge in healthy subjects. Am J Physiol Heart Circ Physiol. 2003; 285: H875–H882.
Marfella R, D’Amico M, Esposito K, Baldi A, Di Filippo C, Siniscalchi M, Sasso FC, Portoghese M, Cirillo F, Cacciapuoti F, Carbonara O, Crescenzi B, Baldi F, Ceriello A, Nicoletti GF, D’Andrea F, Verza M, Coppola L, Rossi F, Giugliano D. The ubiquitin-proteasome system and inflammatory activity in diabetic atherosclerotic plaques: effects of rosiglitazone treatment. Diabetes. 2006; 55: 622–632.
Lee SW, Dai G, Hu Z, Wang X, Du J, Mitch WE. Regulation of muscle protein degradation: coordinated control of apoptotic and ubiquitin-proteasome systems by phosphatidylinositol 3 kinase. J Am Soc Nephrol. 2004; 15: 1537–1545.
Grune T, Blasig IE, Sitte N, Roloff B, Haseloff R, Davies KJ. Peroxynitrite increases the degradation of aconitase and other cellular proteins by proteasome. J Biol Chem. 1998; 273: 10857–10862.
Moens AL, Kass DA. Tetrahydrobiopterin and cardiovascular disease. Arterioscler Thromb Vasc Biol. 2006; 26: 2439–2444.
Diabetes mellitus is one of the leading risk factors for the development of cardiovascular disease. In the past 10 to 15 years, the reason why diabetes mellitus accelerates vascular disease has been the focus of intense study. One reason that is thought to be a key for vascular disease in diabetes mellitus is the effect of diabetes on the function of endothelium. It has been observed in animal models of diabetes mellitus and in diabetic patients that endothelial nitric oxide synthase (eNOS), which is crucial in maintaining endothelium homeostasis, has been transformed from a protective enzyme to a contributor to oxidative stress; this transformation is known as eNOS uncoupling. It is generally agreed that the lack of tetrahydrobiopterin (BH4), the essential cofactor of eNOS, plays a causal role in the development of eNOS uncoupling. However, how a BH4 shortage is developed is poorly understood. In the present study, we have discovered that diabetes mellitus selectively increases the ubiquitination and proteasome degradation of guanosine 5′-triphosphate cyclohydrolase, a rate-limiting enzyme in BH4 de novo synthesis. Furthermore, we found that inhibition of proteasome, an organism whose normal function is cleaning house within a cell, ablated the toxic effects of glucose in cultured endothelial cells and in diabetic mice. Thus, the present study may help to explain why other studies have found that diabetic patients have reduced amounts of nitric oxide with increased production of damaging toxic molecules, a condition that leads to vascular disease. Because proteasome inhibitors increase levels of BH4 but reduce the production of toxic superoxide radicals from the endothelium, proteasome inhibitors, which are currently used in cancer chemotherapy, might be useful tools for treating vascular complications in diabetes mellitus.
The online-only Data Supplement, consisting of a figure, is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.684795/DC1.
Presented in part at the 66th Scientific Sessions of the American Diabetes Association, Washington, DC, June 9–13, 2006.