Published online before print January 24, 2008, doi: 10.1161/CIRCRESAHA.107.156877
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2008 American Heart Association, Inc.
Molecular Medicine |
Extracellular Signal-Regulated Kinase 5 SUMOylation Antagonizes Shear Stress–Induced Antiinflammatory Response and Endothelial Nitric Oxide Synthase Expression in Endothelial Cells
From the Cardiovascular Research Institute (C.-H.W., T.S., C.M., C.Y., J.-i.A.) and Department of Microbiology and Immunology (J.H.L., J.-D.L.), University of Rochester; and Department of Anesthesiology (J.Y.), Columbia University, New York.
Correspondence to Jun-ichi Abe, MD, PhD, Cardiovascular Research Institute, Box 679, 601 Elmwood Ave, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642. E-mail Jun-ichi_Abe{at}urmc.rochester.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Shear stress–induced extracellular signal-regulated kinase (ERK)5 activation and the consequent regulation of Kruppel-like factor 2 and endothelial nitric oxide synthase expression represents one of the antiinflammatory and vascular tone regulatory mechanisms maintaining normal endothelial function. Endothelial dysfunction is a major initiator of atherosclerosis, a vascular pathology often associated with diabetes. Small ubiquitin-like modifier (SUMO) covalently attaches to certain residues of specific target transcription factors and could inhibit its activity. We investigated whether H2O2 and AGE (advanced glycation end products), 2 well-known mediators of diabetes, negatively regulated ERK5 transcriptional activity and laminar flow–induced endothelial nitric oxide synthase expression through ERK5 SUMOylation. H2O2 and AGE induced endogenous ERK5 SUMOylation. In addition, ERK5 SUMOylation was increased in the aortas from diabetic mice. ERK5 transcriptional activity, but not kinase activity, was inhibited by expression of Ubc9 (SUMO E2 conjugase) or PIAS1 (E3 ligase), suggesting the involvement of ERK5 SUMOylation on its transcriptional activity. Point-mutation analyses showed that ERK5 is covalently modified by SUMO at 2 conserved sites, Lys6 and Lys22, and that the SUMOylation defective mutant of ERK5, dominant negative form of Ubc9 (DN-Ubc9), and small interfering RNA PIAS1 reversed H2O2 and AGE–mediated reduction of shear stress–mediated ERK5/myocyte enhancer factor 2 transcriptional activity, as well as promoter activity of Kruppel-like factor 2. Finally, PIAS1 knockdown reversed the inhibitory effect of H2O2 in shear stress–induced Kruppel-like factor 2 and endothelial nitric oxide synthase expression. These data clearly defined SUMOylation-dependent ERK5 transcriptional repression independent of kinase activity and suggested this process as among the molecular mechanisms of diabetes-mediated endothelial dysfunction.
Key Words: ERK5 • SUMOylation • KLF2 • diabetes • shear stress
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Laminar flow–induced extracellular signal-regulated kinase (ERK)5 activation has a critical role in regulating peroxisome proliferator-activated receptor (PPAR)
and Kruppel-like factor (KLF)2 (a recently identified transcriptional inhibitor of endothelial cell [EC] inflammation) and in inhibiting tumor necrosis factor-
–mediated adhesion molecule expression,1 demonstrating an antiinflammatory role of ERK5 activation in ECs. The upstream kinase that phosphorylates ERK5 has been identified as MEK5.2,3 Activation of ERK5 is documented to have an antiapoptotic effect in neuronal and ECs.4,5 Recently we, along with Kasler et al, reported that ERK5 is not only a kinase enzyme but also possesses transcriptional activity.1,6 The NH2-terminal ERK5 kinase domain acts as a negative regulator of these transactivation domains and, when activated, releases the NH2-terminal ERK5 inhibitory effect, resulting in increased COOH-terminal ERK5 transcriptional activity. Both the association of ERK5 with transcriptional factors and a COOH-terminal transcriptional domain of ERK5 are required for myocyte enhancer factor (MEF)2 and PPAR transcriptional activity.1,6
The KLF family is a recently highlighted group of zinc finger transcription factors with important biological roles, including in the vasculature.7 Dekker et al have reported that KLF2 is uniquely induced by laminar flow in the endothelium.8 Overexpression of KLF2 induced endothelial nitric oxide synthase (eNOS) expression as well as inhibited cytokine-mediated adhesion molecule expression.9 Recently, Parmar et al have reported that KLF2 was selectively induced in ECs exposed to a biomechanical stimulus characteristic of atheroprotected regions of the human carotid artery and that this flow-mediated increase in expression occurred via a MEK5/ERK5/MEF2 signaling pathway. In addition, KLF2 induction resulted in the orchestrated regulation of EC transcriptional programs controlling inflammation, thrombosis/hemostasis, vascular tone, and blood vessel development.10 These data collectively implicated laminar flow–mediated KLF2 induction as an antiatherosclerotic and antiinflammatory regulator of EC activation in response to proinflammatory stimuli.7 Both proinflammatory and antiinflammatory stimuli likely modulate PPAR/KLF2 and adhesion molecule (activity or expression) via the transcriptional activity of ERK5. Consistent with such key roles of ERK5 in EC physiology, endothelial-specific ERK5 knockout mice showed a cardiovascular defect attributable to increasing EC apoptosis.4,11
Small ubiquitin-like modifier (SUMO) covalently attaches to certain residues of specific target proteins and alters a number of different protein functions depending on the substrate. Subcellular localization, protein partnering, and DNA binding, and/or transactivation functions of transcription factors are some of the protein functions affected by SUMOylation.12 It is clear that SUMO influences many different biological processes, but particularly important in the present context is the regulation of transcription.13 The SUMOylation pathway is analogous to that of ubiquitination, but SUMO conjugation involves a different set of enzymes. SUMO is activated in an ATP-dependent manner by an E1-activating enzyme that consists of a SAE1(AOS1)-SAE2(UBA2) heterodimer. Activated SUMO is transferred to Ubc9, the E2-conjugating enzyme, and is subsequently attached to the 
-amino group of specific residues in the target protein.14
The formation of reactive oxygen species (ROS) and AGE (advanced glycation end products) is among the major mechanisms that play a role in the pathogenesis of diabetic vascular disease resulting from hyperglycemia.15–18 In our present study, we showed that ROS and AGE lead to ERK5 SUMOylation and inhibition of shear stress–mediated ERK5 transcriptional activity. Inhibition of ERK5 transcriptional activity lead to an inhibition of shear stress–mediated KLF2 promoter activity as well as subsequent KLF2 and eNOS expression in ECs, providing a mechanistic framework to the well-accepted observation that diabetes, through oxidative stress and AGE-dependent signaling, induces endothelial dysfunction, accelerating atherosclerosis formation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Details on reagents, antibodies, plasmid construction, cell culture and transfection, mammalian 1- or 2-hybrid analysis and transfection of cells, flow stimulation, in vitro kinase assay of ERK5, immunoprecipitation (SUMO assay) and Western blot analysis, transfection of the PIAS1 small interfering (si)RNAs, real-time quantitative PCR analysis of KLF2, and statistical analysis are provided in the expanded Materials and Methods section in the online data supplement, available at http://circres.ahajournals.org. All animal procedures were performed with the approval from the University Committee on Animal Resources, University of Rochester.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ERK5 SUMOylation by H2O2 and AGE in Endothelial Cells In Vitro and In Streptozotocin-Treated Diabetic Mice In Vivo
Because it has been reported that SUMOylation could regulate the transcriptional activity of its substrate,13 first we investigated whether H2O2 and AGE induce endogenous ERK5 SUMOylation in ECs in vitro. As shown in Figure 1A, H2O2 significantly increased ERK5 SUMOylation after 30 minutes of H2O2 stimulation, and this ERK5 SUMOylation was maximally increased at 100 µmol/L but decreased at 300 µmol/L, whereas ERK5 phosphorylation was significantly increased at 300 µmol/L H2O2 treatment (Figure 1A, right). We did not observe any significant differences in SUMO2/3 expression between the samples. However, we observed that SUMO modification of proteins around 280 kDa was significantly decreased after 5 minutes of 100 µmol/L H2O2 treatment and also with 300 µmol/L H2O2 treatment (Figure 1A), as described previously.19 These data suggested that ERK5 SUMOylation may have a unique regulatory mechanism different from the general SUMOylation process induced by H2O2. We also tested whether AGE can induce ERK5 SUMOylation. As shown in Figure 1B, 100µg/mL AGE significantly increased ERK5-SUMO conjugation in ECs.
|
To examine the possible pathological role of ERK5 SUMOylation in diabetes mellitus in vivo, diabetes was induced in male FVB mice (8 to 12 weeks old and 25 to 30 g body weight) by intraperitoneal injection of streptozotocin (STZ) (200 mg/kg body weight). As a control, vehicle (0.1 mol/L citrate buffer, pH 4.5) was injected in another group of mice. Tail vein blood glucose samples were measured 6 days after injection to ensure induction of hyperglycemia (STZ injection group: 305.4±59.4 mg/dL; vehicle injection group: 154.9±31.4 mg/dL; mean±SD; P<0.01). At 7 days after injection, mice aortas were isolated from both groups, and ERK5 SUMOylation assay was performed with immunoprecipitation of ERK5 and immunoblotted with anti-SUMO2/3 antibody. As shown in Figure 2A and 2B, we found that ERK5 SUMOylation was increased in STZ-injected mice compared to vehicle-injected mice. These data suggested the possible pathological role of ERK5 SUMOylation not only in vitro but also in vivo.
|
H2O2 and AGE Increased ERK5 Phosphorylation but Inhibited ERK5 Transcriptional Activity
To determine the effects of ROS and AGE on ERK5, first we determined ERK5 phosphorylation by immunoblotting with anti–phospho-ERK5 antibody. ERK5 phosphorylation is directly related to its kinase activation,20,21 and, as we reported previously,22 H2O2 significantly increased ERK5 phosphorylation. Similarly, AGE increased ERK5 phosphorylation (Figure 3A and 3B). ERK5 kinase activation and its autophosphorylation are associated with a mobility shift (Figure 3A and 3B, second from the top). Because ERK5 is not only a kinase but also possesses strong transcriptional activity and increases MEF2 activity,1,6 we investigated the effect of H2O2 and AGE on ERK5 transcriptional activity using a Gal4-ERK5 construct. In contrast to ERK5 phosphorylation, we could not detect any induction of ERK5 transcriptional activity by H2O2 and AGE. In fact, we found that H2O2 and AGE concentration-dependently inhibited CA-MEK5–induced ERK5 transcriptional activity (Figure 3C and 3D). On the other hand, H2O2 and AGE did not affect the CA-MEK5–induced ERK5 phosphorylation (Figure 3B). Transfection with CA-MEK5 induced ERK5 phosphorylation and kinase activation, as well as ERK5 transcriptional activity. As we and others previously reported,1,6 ERK5 phosphorylation and kinase activation increases ERK5 transcriptional activity by releasing the inhibitory effect of NH2-terminal ERK5 kinase domain on the COOH-terminal ERK5 transcriptional activity. However, although H2O2 and AGE induced ERK5 phosphorylation and kinase activation, H2O2 and AGE inhibited CA-MEK5–mediated ERK5 transcriptional activity, suggesting a differential regulation of ERK5 phosphorylation and transcriptional activity by H2O2 and AGE in contrast to stimulation by CA-MEK5. We found a similar inhibitory effect of H2O2 and AGE on ERK5 transcriptional activity in human umbilical vein ECs (HUVECs) as we describe in Figures 4 and 5
.
|
|
|
SUMOylation of ERK5
To gain further insight into the function and regulation of the ERK5 transcriptional activity, we inspected the amino acid sequences to identify motifs that may suggest potential covalent modification sites. The in silico analysis revealed a putative SUMOylation motif that is conserved among ERK5 proteins from human, mouse, and rat (supplemental Figure IA). The motifs (K6 and K22) are located at the divergent N-terminal region close to the ATP-binding site and contain a cytoplasmic targeting region. We next investigated whether ERK5 can be SUMOylated. In vivo SUMOylation assays with HA-tagged SUMO proteins were performed. An expression plasmid for Flag-tagged ERK5 was cotransfected into CHO cells along with expression plasmids for HA-tagged SUMO1 and SUMO3 with or without Ubc9. SUMO1 protein exhibits 44% sequence identity with SUMO2 and SUMO3 protein, whereas SUMO2 and SUMO3 protein share 86% sequence identity.23 In addition, it has been reported that SUMO1 and SUMO2/3 have different substrate specificity.23–25 Therefore, in this study, we selected SUMO1 and SUMO3 for further examination. After transfection, cell lysates were immunoprecipitated by the anti-flag (Flag-ERK5) antibody, and immunoprecipitates were immunoblotted with the anti-HA (HA-SUMO) antibody. As shown in supplemental Figure IB, SUMO conjugation of 70- to 280-kDa proteins was increased by both SUMO1/3 and Ubc9 expression (third panel from the top), and Flag-ERK5 was also significantly modified by both SUMO1 and SUMO3. In addition, immunoblotting with anti-ERK5 antibody after immunoprecipitation by anti-Flag antibody showed a significant band shift of ERK5 after cotransfection with SUMO1 or -3 and Ubc9 (supplemental Figure IB, second from the top, lanes 2 and 4). Of note, we did not find any increase in ERK5 phosphorylation or kinase activation induced by Ubc9 (supplemental Figure IVA and IVB), although we found a significant band shift by Ubc9 transfection, suggesting that the ERK5 band shift may represent both phosphorylation and SUMOylation. To determine the role of the protein inhibitor of activated STAT 1 (PIAS1), the SUMO E3-ligase enzyme, in ERK5 SUMOylation, an expression plasmid for Flag-tagged ERK5 was cotransfected into CHO cells along with expression plasmids for PIAS1 with or without DN-Ubc9 (supplemental Figure IC). Overexpression of PIAS1 induced ERK5 SUMOylation, and DN-Ubc9 inhibited it (supplemental Figure IC, top, lanes 3 and 4), indicating the involvement of PIAS1 and Ubc9 in ERK5 SUMOylation as well.
SUMO3 Modification of ERK5 on K6 and K22
Because Lys6 and Lys22 of ERK5 are the putative SUMOylation sites, these residues were replaced by arginine to create the point mutants K6R and K22R and the double-mutant K6/22R. These mutants were expressed in CHO cells and analyzed for conjugation with HA-SUMO3. As shown in supplemental Figure IIA, K6R and K22R demonstrated decreased SUMO modification and, unlike wild-type ERK5, K6/22R was not modified by HA-SUMO3, suggesting that Lys6 and -22 were the SUMOylation sites of ERK5. To confirm the ERK5 SUMOylation at K6 and K22, cell lysates were conversely immunoprecipitated by anti-HA (HA-SUMO3) and immunoblotted with anti-ERK5 antibody. As shown in supplemental Figure IIB, SUMOylated ERK5 was detected in ERK5 wild-type transfected cells, but ERK5 SUMOylation was significantly decreased in ERK5 mutants, confirming that both K6 and K22 of ERK5 are SUMOylated. As seen in the longer exposure image of the same blot, the K6/22R mutation did not completely abolish ERK5 SUMOylation, suggesting the possible existence of other minor SUMOylation sites in addition to K6 and K22.
Regulation of the Transcriptional Activities of ERK5 and MEF2 by ERK5 SUMOylation
SUMOylation has been shown to inhibit transcriptional activities of other transcription factors,26 and because K6 and K22 are located in the N-terminal inhibitory domain of ERK5 transcriptional activity,1 we investigated whether SUMOylation affected transcriptional activity. Cells were cotransfected with Ubc9 and PIAS1 with or without CA-MEK5 and ERK5 transcriptional activity evaluated by a reporter gene assay as described in Methods.
As shown in supplemental Figure IIIA, CA-MEK5–induced ERK5 transcriptional activity was inhibited by cotransfection with Ubc9 or PIAS1, suggesting an inhibitory effect of ERK5 SUMOylation on its transcriptional activity. To substantiate this, we compared the ERK5 transcriptional activity of the wild type with K6/22R mutant using fusion proteins with the DNA-binding domain of Gal4 cotransfected with the pG5-luc reporter. Although there was a slightly lower expression level of the K6/22R mutant compared with wild type (supplemental Figure IIID), the transcriptional activity of ERK5 K6/22R mutant was greater than the wild type (supplemental Figure III). In addition, the ERK5-mediated MEF2 transcriptional activity was enhanced in the K6/22R mutant (supplemental Figure IIIC), suggesting that the K6 and K22 sites of ERK5 negatively regulate ERK5 transcriptional activity as well as subsequent MEF2 transcriptional activity.
Next, to determine the role of SUMOylation of K6 and K22 sites, we compared the reduction of ERK5 transcription induced by Ubc9 or PIAS1 between wild type and the K6/22R mutant. As shown in supplemental Figure IIIE, the reduction of ERK5 wild-type transcriptional activity by Ubc9 and PIAS1 was significantly less in the K6/22R mutant. These data suggested that SUMOylation on K6 and K22 sites inhibited ERK5 transcriptional activity.
ERK5 SUMOylation Did Not Affect ERK5 Phosphorylation and Kinase Activity but Inhibited ERK5 and Coactivator SRC-1 Interaction
Because ERK5 transcriptional activity was inhibited by SUMOylation, we investigated whether Ubc9 could inhibit ERK5 phosphorylation and kinase activation. As shown in supplemental Figure IVA and IVB, we did not observe any difference in ERK5 phosphorylation and kinase activity between Ubc9-overexpressed and vector-transfected cells. In addition, we did not find any difference in ERK5 phosphorylation between the ERK5 wild type and K6/22R mutant (supplemental Figure IVC). These data suggested that ERK5 SUMOylation did not change ERK5 phosphorylation and kinase activation.
The reduction of ERK5 transcriptional activity induced by ERK5 SUMOylation could be attributable to the increase of ERK5 interaction with repressors or the decrease of ERK5 interaction with coactivators.27,28 First, we investigated whether ERK5 can associate with the corepressors SMRT and NCoR1 or coactivator SRC-1 by using a mammalian 2-hybrid assay. No ERK5 association with the corepressors SMRT and NCoR1 was found, but ERK5 associated with SRC-1, and this association was significantly increased by CA-MEK5 cotransfection (supplemental Figure IVD). Next, to determine the involvement of ERK5 SUMOylation on ERK5–SRC-1 association, we cotransfected with PIAS1 and Ubc9 or DN-Ubc9 and determined the ERK5–SRC-1 association. As shown in supplemental Figure IVE and IVF, PIAS1 or Ubc9 transfection decreased ERK5–SRC-1 association. In contrast, DN-Ubc9 transfection increased ERK5–SRC-1 association, suggesting that ERK5 SUMOylation may inhibit ERK5–SRC-1 interaction and decreases ERK5 transcriptional activity. In addition, the inhibition of ERK5 interaction with SRC-1 was significantly decreased in the K6/22R mutant, also suggesting the mechanistic role of ERK5 SUMOylation in regulating ERK5 transcriptional activity. Further investigation is required to determine the exact role of ERK5–SRC-1 interaction in regulating ERK5 transcriptional activity.
H2O2 and AGE Inhibit ERK5 Transcriptional Activity via ERK5 SUMOylation
To investigate whether ERK5 SUMOylation is important in H2O2- or AGE-mediated reduction of ERK5 transcriptional activity, we first compared the reduction of ERK5 transcriptional activity by H2O2 (100 µmol/L) between the wild-type and K6/22R mutant in HUVECs. As shown in Figure 4A, H2O2 decreased ERK5 transcriptional activity by 
41%, but the H2O2-mediated inhibition of ERK5 transcriptional activity was significantly less in the K6/22R mutant. To further confirm the involvement of ERK5 SUMOylation, we tested whether transfection of DN-Ubc9 might inhibit the H2O2-mediated reduction of ERK5 transcriptional activity. DN-Ubc9 prevented H2O2-mediated reduction of ERK5 transcriptional activity (Figure 4B). We also observed a similar regulation of ERK5 transcriptional activity via ERK5 SUMOylation induced by H2O2 or AGE in bovine aortic ECs (BAECs) (supplemental Figure V). Furthermore, instead of using CA-MEK5 transfection to increase ERK5 transcriptional activity, we also investigated whether H2O2 or AGE could inhibit ERK5 activation induced by laminar flow. We transfected Gal4-tagged wild-type or K6/22R ERK5 constructs and stimulated the cells with vehicle or H2O2 (100 µmol/L) and BSA or AGE-BSA (100 µg/mL) 24 hours after transfection. One hour after H2O2 or AGE stimulation, HUVECs were exposed to flow or static conditions, and luciferase activity was assayed 24 hours later. H2O2 and AGE significantly inhibited ERK5 wild-type transcriptional activity, but the H2O2 and AGE-mediated reduction of ERK5 transcriptional activity was less in the K6/22R mutant (Figure VA and VB).
The importance of MEF2/KLF2 in the regulation of endothelial inflammation, thrombosis/hemostasis, vascular tone, and blood vessel development has been reported.9,29,30 Therefore, we investigated whether ERK5 SUMOylation was involved in the H2O2- or AGE-mediated reduction of MEF2 transcriptional activity and the subsequent KLF2 promoter activity. As shown in supplemental Figure VIA and VIB, both H2O2 and AGE significantly decreased the MEF2 transcriptional activity in wild-type ERK5 transfected cells, but this inhibitory effect was less in the mutant K6/22R ERK5-transfected cells. Although it has been reported that MEF2 SUMOylation regulates its own transcriptional activity, we found that the K6/22R ERK5 mutant almost completely inhibited the H2O2- and AGE-mediated decrease of ERK5 transcriptional activity, suggesting that ERK5 SUMOylation, rather than MEF2 SUMOylation, may be the main contributor to the reduction of MEF2 transcriptional activity by these agents in ECs. Finally, we determined the involvement of ERK5 SUMOylation on the regulation of KLF2 promoter activity, which is a downstream target of MEF2 activity.31 H2O2 significantly decreased the KLF2 promoter activity in ERK5 wild-type transfected cells (supplemental Figure VIC). In contrast, the inhibition of KLF2 promoter activity induced by H2O2 was completely abolished in the K6/22R ERK5 mutant transfected cells. These data also suggested that the inhibition of ERK5/MEF2/KLF2 pathway by H2O2 is mediated by ERK5 SUMOylation.
ERK5 SUMOylation induced by H2O2 and AGE significantly attenuates KLF2 and subsequent eNOS expression induced by atheroprotective flow in ECs.
Previous studies have shown that flow-mediated ERK5/MEF2/KLF2 induction leads to the upregulation of eNOS and the inhibition of endothelial inflammation.10 Because we found that H2O2 and AGE induced ERK5 SUMOylation and inhibited the ERK5/MEF2/KLF2 signaling pathway, we investigated whether H2O2 or AGE-mediated ERK5 SUMOylation was involved in the flow-mediated KLF2 induction and subsequent eNOS expression. As shown in Figure 6A, the KLF2 promoter activity was significantly increased by flow in ERK5 wild-type transfected ECs. Next, we cotransfected wild-type or mutant K6/22R ERK5 constructs with the KLF2 promoter reporter gene and stimulated the cells with or without H2O2 (100 µmol/L) 24 hours after transfection. One hour after H2O2 stimulation, ECs were exposed to flow or static conditions, and luciferase activity was assayed 24 hours after the flow stimulation. H2O2 significantly inhibited the KLF2 promoter activity in wild-type ERK5 transfected cells. In contrast, mutant K6/22R ERK5 significantly inhibited H2O2-mediated reduction of KLF2 promoter activity (Figure 6A), and a similar tendency was observed in AGE-mediated inhibition of KLF2 promoter activity (Figure 6B). To further examine the involvement of ERK5 SUMOylation in this process, we used PIAS1 siRNA to ablate the PIAS1 expression. PIAS1 siRNA caused a specific reduction in the expression of endogenous PIAS1 (Figure 6C, right). As shown in Figure 6C, the ablation of PIAS1 prevented H2O2-mediated reduction of KLF2 promoter activity in HUVECs.
|
We confirmed these results at the expression level for KLF2 and eNOS (Figure 7). The flow-mediated increase in KLF2 (mRNA) and eNOS (protein) expression were significantly decreased by H2O2 pretreatment (Figure 7A and 7B). However, the reduction of PIAS1 expression by PIAS1 siRNA treatment abolished this H2O2-mediated inhibition of KLF2 and eNOS expression induced by flow. Together, these data demonstrated that ERK5 SUMOylation by H2O2 is a critical regulator on laminar flow–mediated KLF2 as well as on subsequent eNOS expression in ECs.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Endothelial inflammation is among the major initiators of atherosclerosis, and diabetes significantly affects this process. However, what is lacking is a plausible relationship between diabetes and any of the known regulators of endothelial inflammation that play a significant role in accelerating atherosclerosis formation in diabetes. Recently, we and others found that laminar flow, but not disturbed flow, acts as an antiinflammatory modulator by increasing ERK5 and PPAR
transcriptional activity and decreasing adhesion molecule expression in ECs. Laminar flow–induced ERK5 activation has a critical role in regulating PPAR activation and KLF2, a recently identified transcriptional inhibitor of endothelial inflammation, as well as inhibiting tumor necrosis factor-–mediated adhesion molecule expression, suggesting the critical antiinflammatory role of ERK5 activation in ECs.1,10 Increased expression of adhesion molecules and inflammation in diabetes has been reported in vitro and in vivo.32,33 However, the molecular mechanism of this diabetes-mediated inflammatory responses and adhesion molecule induction remains unclear. In the present study, we found that H2O2 and AGE significantly inhibited ERK5/MEF2 transcriptional activity as well as the subsequent shear stress–mediated KLF2 promoter activity and KLF2 and eNOS expression via ERK5 SUMOylation, although H2O2 and AGE increased ERK5 phosphorylation and kinase activation (Figure 8). In addition, we found that ERK5 SUMOylation was increased in the vessels of diabetic mice in vivo. We propose that inhibition of ERK5 SUMOylation may be a new therapeutic target for the treatment of diabetes-mediated endothelial dysfunction and inflammation.
|
It remains unclear how H2O2 and AGE increase ERK5 SUMOylation, but several previous reports suggested the possible involvement of other posttranslational modification such as phosphorylation or association with other adapter molecules in the regulation of SUMOylation. For example, Gregoire and Yang have reported that class IIa histone deacetylases (HDACs) stimulate the SUMOylation of MEF2.34 Interestingly, this stimulation of MEF2 SUMOylation was dependent on residues 118 to 488 at MEF2D but not the C-terminal catalytic domain of HDAC4, suggesting that the deacetylase activity of HDAC4 is not required. This may provide another possible mechanism by which HDACs can inhibit transcriptional activity via increasing SUMOylation. Furthermore, they found that ERK5, which is known to phosphorylate Ser179 of MEF2D, negatively regulated MEF2 SUMOylation.34 Although particular phosphorylation sites or HDAC association with ERK5 induced by H2O2 and AGE is unknown, it is well known that both H2O2 and AGE can activate many kinases that may contribute to the regulation of ERK5 SUMOylation. In the present study, we focused on ERK5-mediated KLF2 and eNOS expression, which also mediates the antiinflammatory effect of shear stress.10 ERK5 SUMOylation inhibited ERK5 transcriptional activity, indicating that ERK5 SUMOylation can inhibit both ERK5/MEF2 and PPAR and promote a strong proinflammatory and atherogenic effect in ECs.
An expanded Discussion section is in the online data supplement.
|
Acknowledgments |
|---|
Sources of Funding
This work was supported by America Heart Association Postdoctoral Fellowship 0625957T (to C.-H.W.) and NIH grants GM-071485 and HL-077789 (to J.-i.A.) and HL-077789 and GM-071485 (to J.Y.). J.-i.A. and J.Y. are recipients of Established Investigator Awards from the American Heart Association (0740013N and 0740021N).
Disclosures
None.
|
Footnotes |
|---|
Original received May 29, 2007; revision received January 10, 2008; accepted January 16, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Akaike M, Che W, Marmarosh NL, Ohta S, Osawa M, Ding B, Berk BC, Yan C, Abe J. The hinge-helix 1 region of peroxisome proliferator-activated receptor gamma1 (PPARgamma1) mediates interaction with extracellular signal-regulated kinase 5 and PPARgamma1 transcriptional activation: involvement in flow-induced PPARgamma activation in endothelial cells. Mol Cell Biol. 2004; 24: 8691–8704.
2. Zhou G, Bao ZQ, Dixon JE. Components of a new human protein kinase signal transduction pathway. J Biol Chem. 1995; 270: 12665–12669.
3. Lee JD, Ulevitch RJ, Han J. Primary structure of BMK1: a new mammalian MAP kinase. Biochem Biophys Res Commun. 1995; 213: 715–724.[CrossRef][Medline] [Order article via Infotrieve]
4. Hayashi M, Kim SW, Imanaka-Yoshida K, Yoshida T, Abel ED, Eliceiri B, Yang Y, Ulevitch RJ, Lee JD. Targeted deletion of BMK1/ERK5 in adult mice perturbs vascular integrity and leads to endothelial failure. J Clin Invest. 2004; 113: 1138–1148.[CrossRef][Medline] [Order article via Infotrieve]
5. Liu L, Cundiff P, Abel G, Wang Y, Faigle R, Sakagami H, Xu M, Xia Z. Extracellular signal-regulated kinase (ERK) 5 is necessary and sufficient to specify cortical neuronal fate. Proc Natl Acad Sci U S A. 2006; 103: 9697–9702.
6. Kasler HG, Victoria J, Duramad O, Winoto A. ERK5 is a novel type of mitogen-activated protein kinase containing a transcriptional activation domain. Mol Cell Biol. 2000; 20: 8382–8389.
7. Suzuki T, Aizawa K, Matsumura T, Nagai R. Vascular implications of the Kruppel-like family of transcription factors. Arterioscler Thromb Vasc Biol. 2005; 25: 1135–1141.
8. Dekker RJ, van Soest S, Fontijn RD, Salamanca S, de Groot PG, VanBavel E, Pannekoek H, Horrevoets AJ. Prolonged fluid shear stress induces a distinct set of endothelial cell genes, most specifically lung Kruppel-like factor (KLF2). Blood. 2002; 100: 1689–1698.
9. SenBanerjee S, Lin Z, Atkins GB, Greif DM, Rao RM, Kumar A, Feinberg MW, Chen Z, Simon DI, Luscinskas FW, Michel TM, Gimbrone MA Jr, Garcia-Cardena G, Jain MK. KLF2 Is a novel transcriptional regulator of endothelial proinflammatory activation. J Exp Med. 2004; 199: 1305–1315.
10. Parmar KM, Larman HB, Dai G, Zhang Y, Wang ET, Moorthy SN, Kratz JR, Lin Z, Jain MK, Gimbrone MA Jr, Garcia-Cardena G. Integration of flow-dependent endothelial phenotypes by Kruppel-like factor 2. J Clin Invest. 2006; 116: 49–58.[CrossRef][Medline] [Order article via Infotrieve]
11. Sohn SJ, Sarvis BK, Cado D, Winoto A. ERK5 MAPK regulates embryonic angiogenesis and acts as a hypoxia-sensitive repressor of vascular endothelial growth factor expression. J Biol Chem. 2002; 277: 43344–43351.
12. Hilgarth RS, Murphy LA, Skaggs HS, Wilkerson DC, Xing H, Sarge KD. Regulation and function of SUMO modification. J Biol Chem. 2004; 279: 53899–53902.
13. Johnson ES. Protein modification by SUMO. Annu Rev Biochem. 2004; 73: 355–382.[CrossRef][Medline] [Order article via Infotrieve]
14. Melchior F. SUMO–nonclassical ubiquitin. Annu Rev Cell Dev Biol. 2000; 16: 591–626.[CrossRef][Medline] [Order article via Infotrieve]
15. Hudson BI, Wendt T, Bucciarelli LG, Rong LL, Naka Y, Yan SF, Schmidt AM. Diabetic vascular disease: it’s all the RAGE. Antioxid Redox Signal. 2005; 7: 1588–1600.[CrossRef][Medline] [Order article via Infotrieve]
16. Rask-Madsen C, King GL. Mechanisms of Disease: endothelial dysfunction in insulin resistance and diabetes. Nat Clin Pract Endocrinol Metab. 2007; 3: 46–56.[CrossRef][Medline] [Order article via Infotrieve]
17. Ting HH, Timimi FK, Boles KS, Creager SJ, Ganz P, Creager MA. Vitamin C improves endothelium-dependent vasodilation in patients with non-insulin-dependent diabetes mellitus. J Clin Invest. 1996; 97: 22–28.[Medline] [Order article via Infotrieve]
18. Park L, Raman KG, Lee KJ, Lu Y, Ferran LJ Jr, Chow WS, Stern D, Schmidt AM. Suppression of accelerated diabetic atherosclerosis by the soluble receptor for advanced glycation endproducts. Nat Med. 1998; 4: 1025–1031.[CrossRef][Medline] [Order article via Infotrieve]
19. Bossis G, Melchior F. Regulation of SUMOylation by reversible oxidation of SUMO conjugating enzymes. Mol Cell. 2006; 21: 349–357.[CrossRef][Medline] [Order article via Infotrieve]
20. Abe J, Takahashi M, Ishida M, Lee JD, Berk BC. c-Src is required for oxidative stress-mediated activation of big mitogen-activated protein kinase 1. J Biol Chem. 1997; 272: 20389–20394.
21. Zheng Q, Yin G, Yan C, Cavet ME, Berk BC. 14-3-3beta binds to big mitogen activated protein kinase 1 (BMK1/ERK5) and regulates BMK1 function. J Biol Chem. 2004; 279: 8787–8791.
22. Abe J, Kusuhara M, Ulevitch RJ, Berk BC, Lee JD. Big mitogen-activated protein kinase 1 (BMK1) is a redox-sensitive kinase. J Biol Chem. 1996; 271: 16586–16590.
23. Su HL, Li SS. Molecular features of human ubiquitin-like SUMO genes and their encoded proteins. Gene. 2002; 296: 65–73.[CrossRef][Medline] [Order article via Infotrieve]
24. Hecker CM, Rabiller M, Haglund K, Bayer P, Dikic I. Specification of SUMO1- and SUMO2-interacting motifs. J Biol Chem. 2006; 281: 16117–16127.
25. Saitoh H, Hinchey J. Functional heterogeneity of small ubiquitin-related protein modifiers SUMO-1 versus SUMO-2/3. J Biol Chem. 2000; 275: 6252–6258.
26. Verger A, Perdomo J, Crossley M. Modification with SUMO. A role in transcriptional regulation. EMBO Rep. 2003; 4: 137–142.[CrossRef][Medline] [Order article via Infotrieve]
27. Girdwood D, Bumpass D, Vaughan OA, Thain A, Anderson LA, Snowden AW, Garcia-Wilson E, Perkins ND, Hay RT. P300 transcriptional repression is mediated by SUMO modification. Mol Cell. 2003; 11: 1043–1054.[CrossRef][Medline] [Order article via Infotrieve]
28. Goodman RH, Smolik S. CBP/p300 in cell growth, transformation, and development. Genes Dev. 2000; 14: 1553–1577.
29. Nuez B, Michalovich D, Bygrave A, Ploemacher R, Grosveld F. Defective haematopoiesis in fetal liver resulting from inactivation of the EKLF gene. Nature. 1995; 375: 316–318.[CrossRef][Medline] [Order article via Infotrieve]
30. Kuo CT, Veselits ML, Leiden JM. LKLF: a transcriptional regulator of single-positive T cell quiescence and survival. Science. 1997; 277: 1986–1990.
31. Kumar A, Lin Z, SenBanerjee S, Jain MK. Tumor necrosis factor alpha-mediated reduction of KLF2 is due to inhibition of MEF2 by NF-kappaB and histone deacetylases. Mol Cell Biol. 2005; 25: 5893–5903.
32. Rohrer L, Hersberger M, von Eckardstein A. High density lipoproteins in the intersection of diabetes mellitus, inflammation and cardiovascular disease. Curr Opin Lipidol. 2004; 15: 269–278.[CrossRef][Medline] [Order article via Infotrieve]
33. Alexander RW. Atherosclerosis as disease of redox-sensitive genes. Trans Am Clin Climatol Assoc. 1998; 109: 129–145.[Medline] [Order article via Infotrieve]
34. Gregoire S, Yang XJ. Association with class IIa histone deacetylases upregulates the sumoylation of MEF2 transcription factors. Mol Cell Biol. 2005; 25: 2273–2287.
Related Article:
Circ. Res. 2008 102: 510-512.
This article has been cited by other articles:
 |
|
 |
|
  I. R. Buschmann, K. Lehmann, F. Le Noble, and on behalf of Art.Net. Physics Meets Molecules: Is Modulation of Shear Stress the Link to Vascular Prevention? Circ. Res., March 14, 2008; 102(5): 510 - 512. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||