Insulin-Like Growth Factor-1 Enhances Inflammatory Responses in Endothelial Cells
Role of Gab1 and MEKK3 in TNF-α–Induced c-Jun and NF-κB Activation and Adhesion Molecule Expression
Insulin-like growth factor (IGF)-1 and the type I IGF-1 receptor are important regulators of vascular function that may contribute to cardiovascular disease. We hypothesized that IGF-1 causes endothelial cell dysfunction and expression of neutrophil and monocyte adhesion molecules by enhancing pro-inflammatory cytokine signal transduction. Long-term IGF-1 treatment of endothelial cells potentiated c-Jun and nuclear factor NF-κB activation by tumor necrosis factor (TNF)-α and enhanced TNF-α–mediated adhesion molecule expression. In response to IGF-1 treatment, the expression of kinases in the c-Jun/c-Jun NH2-terminal kinase signaling pathway (MEKK1, MEK4, and JNK1/2) was unchanged, but expressions of insulin receptor substrate-1 and Grb2-associated binder-1 (Gab1) were significantly decreased. Because Gab1 is involved in both c-Jun and NF-κB activation by TNF-α, we focused on Gab1-dependent signaling. Gab1 inhibited c-Jun and NF-κB transcriptional activation by TNF-α. Interestingly, Gab1 inhibited c-Jun transcriptional activity induced by MEKK3 but not MEKK1 and MEK4. Gab1 associated with MEKK3, and a catalytically inactive form of MEKK3 inhibited TNF-α–induced c-Jun and NF-κB transcriptional activation, suggesting a critical role for Gab1 and MEKK3 in TNF-α signaling. These data demonstrate that Gab1 and MEKK3 play important roles in endothelial cell inflammation via regulating the activation of c-Jun and NF-κB. Furthermore, the IGF-1–mediated downregulation of Gab1 expression represents a novel mechanism to promote vascular inflammation and atherosclerosis.
- insulin-like growth factor-1
- signal transduction
- Grb2-associated binder-1
- tumor necrosis factor-&agr
- vascular inflammation
There is increasing support for the idea that inflammation plays a major role in the development of atherosclerosis.1 It is well known that inflammatory stimuli increase adhesion molecule expression at the initial sites of atherosclerosis. The adhesion of mononuclear leukocytes to the intact endothelium occurs very early in atherogenesis.1 In fact, it has been reported that an important feature of endothelial dysfunction, in the context of hypercholesterolemia, is the increased adhesion of mononuclear leukocytes to the endothelial surface at sites of subsequent fatty streaks and foam cell accumulation.1,2⇓
Insulin-like growth factor (IGF)-1 is a polypeptide growth factor that binds to the type I IGF-1 receptor present on many cell types, including vascular smooth muscle cells (VSMCs) and endothelial cells.3,4⇓ It has been reported that IGF-1 plays a role in cellular growth and survival in cardiovascular tissue, especially in pathological states.4 Serum and tissue concentrations of IGF-1 are highly regulated by multiple IGF-1–binding proteins. However, it is estimated that IGF-1 circulates at 50 to 200 ng/mL, making it potentially more important than insulin, which circulates at 0.5 to 6 ng/mL. Recently, Anwar et al5 have reported that tumor necrosis factor (TNF)-α, a cytokine that is upregulated in atherosclerotic plaques, decreases IGF-1 expression in VSMCs. Because IGF-1 is important for the growth and survival of VSMCs, these authors have suggested that IGF-1 downregulation by TNF-α plays a role in acute coronary syndromes by decreasing VSMC viability and promoting plaque instability.5 Conversely, in the present study, we propose that IGF-1 augments proinflammatory events that are induced by TNF-α in endothelial cells and that promote atherogenesis.
We show that long-term IGF-1 treatment enhances c-Jun and nuclear factor NF-κB activation, which is regulated by TNF-α, and enhances TNF-α–induced adhesion molecule expression in endothelial cells, leading to the increased adhesion of monocytes. IGF-1 treatment did not change the expression of kinases involved in the c-Jun and NF-κB signaling pathway (eg, MEKK1, MEK4, and JNK1/2) but decreased insulin receptor substrate (IRS)-1 and Grb2-associated binder-1 (Gab1) expression. A critical role for Gab1 has been found, inasmuch as inhibiting c-Jun and NF-κB signaling by overexpression of Gab1 decreased endothelial intercellular adhesion molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1, and E-selectin expression. These results provide the first evidence that long-term IGF-1 treatment leads to hypersensitivity to TNF-α–mediated signaling events via reduction in Gab1 expression.
Materials and Methods
Cell Culture and Materials
Bovine aortic endothelial cells (BAECs) were obtained from Cell Systems and maintained in culture in medium 199 (Cellgro) supplemented with 10% FBS, as described previously.6 Human umbilical vein endothelial cells (HUVECs) were isolated and maintained in 20% FBS/RPMI 1640 medium, as described previously.7 The HUVECs were plated onto gelatin-coated culture dishes or plates and were studied before cell confluence between passages 3 and 9. U937 cells were obtained from American Type Culture Collection and maintained in RPMI-1640 medium as previously described.8 Human recombinant IGF-1 was from Calbiochem. Human recombinant TNF-α was purchased from GIBCO.
Plasmids and Transfection
pcDNA3 HA-Gab1 was a kind gift from Dr Hirano (Osaka University Medical School, Japan),9 and we subcloned Gab1 into pcDNA3.1/His C vector (Invitrogen), which contains Xpress and His tags. pBTM-IRS-1 was a kind gift from Dr White (Harvard Medical School, Boston, Mass),10 and a constitutively active mutant form of MEK4 (pCMV CA-MEK4), in which Ser220 and Thr224 were mutated to glutamic acid and aspartic acid, was from Dr Morrison (Washington University, School of Medicine, St Louis, Mo).11 pRK5-MEKK3 (wild type), pRK5-CA-MEKK3 (constitutively active [CA] form by deletion of the first 11 amino acids), and DN-MEKK3 (dominant-negative form [DN] by changing the ATP binding site from Lys391 to Trp391) plasmids12 were subcloned into pCMV-tag2 (Stratagene), with the addition of Flag tag (Sigma Chemical Co). pFR-Luc, pFA2-c-Jun, pFC-dbd, pNF-κB-Luc, and pFC-MEKK1 were from Stratagene. For transient expression experiments, BAECs and HUVECs were transfected with Lipofectamine Plus (GIBCO-BRL), as described previously.13 We determined the transfection efficiency with pcDNA3.1-LacZ transfection and by X-Gal staining. The transfection efficiency was 5% to 10% in HUVECs and 20% to 25% in BAECs.
Immunoprecipitation and Western Blot Analysis
After treatment with reagents, the cells were washed with PBS and harvested in 0.5 mL lysis buffer as described previously.13 Immunoprecipitation was performed as described previously with mouse anti-His (6×His-Gly epitope, Invitrogen) or anti-HA (F-7) (Santa Cruz) antibody.14 Western blot analysis was performed as previously described.13 In brief, the blots were incubated for 4 hours at room temperature with the following antibodies: MEKK1, MEK4, 14-3-3β, IRS-1, and IRS-2 (Santa Cruz); MEKK3 (Stressgene); Gab1 (U.B.I.); c-Jun, phospho-c-Jun, and phospho-MEK4 (Cell Signaling); Xpress (Invitrogen); or with Flag (Sigma) antibodies, followed by incubation with horseradish peroxidase–conjugated secondary antibody (Amersham). Immunoreactive bands were visualized by using enhanced chemiluminescence (ECL, Amersham).
JNK Activity Assay
JNK activity was measured with a commercially available kit based on phosphorylation of recombinant c-Jun (New England Biolabs), as described previously.13
PathDetect In Vivo Signal Transduction Pathway Reporting System
c-Jun activity was assayed by using PathDetect Signal Transduction Pathway trans-Reporting Systems (Stratagene), as described previously.15 NF-κB activity was measured by PathDetect Signal Transduction Pathway cis-Reporting Systems (Stratagene). Cells were cotransfected with pNF-κB-Luc reporter plasmid and pRL-PCMV and with other plasmids as indicated in the figures. After stimulation with a reagent, the cells were assessed for luciferase activity.
Preparation of Recombinant Proteins and Coimmunoprecipitation (TNT-Coupled Reticulocyte Systems Assay)
Recombinant His-Gab1, CA-MEKK3, and DN-MEKK3 were expressed by the TNT-Coupled Reticulocyte Lysate System (Promega) using pcDNA3.1 His-Gab1, pRK5 CA-MEKK3, and pRK5 DN-MEKK3, incorporating [35S]methionine, as previously described.16
Cell Surface Immunoassay
To detect the expression of adhesion molecules, ELISA was performed as previously described.17 In brief, after 2% gelatin was coated in flat-bottomed 96-well plates, HUVECs were plated and incubated overnight. The cells were transfected with MEKK3 plasmid or empty vector, as described above. Forty-eight hours after transfection, HUVECs were washed with HBSS and fixed. Anti–ICAM-1, anti–VCAM, or anti–E-selectin antibody (Chemicon) was added to the wells. After horseradish peroxidase–conjugated goat anti-mouse IgG was added, the cells were incubated with o-phenylenediamine dihydrochloride in phosphate-citrate buffer. Absorbance at 490 nm was determined by a 1420 Multilabel counter (Victor2, Wallac).
HUVECs were plated on 2% gelatin–coated 12-well plates and cultured to confluence in 20% FBS containing RPMI 1640 medium. The cells were incubated in 10 ng/mL IGF-1 for 16 hours and then were treated with 5 ng/mL TNF-α for 8 hours. Human U937 cells were washed 3 times with serum-free RPMI 1640 medium and suspended in the serum-free RPMI 1640 medium. Approximately 1 mL of the cells (20 000 cells/mL) were put into the wells and incubated for 20 minutes. Then unfixed cells in the wells were washed out 3 times with serum-free RPMI 1640 medium. The adherent cells were counted in 5 randomly selected optical fields in each well, as previously described.18 Phase-contrast microphotographs of the cells in plates were taken under a microscope (IMT-2, Olympus) with an SC 35 Type 12 camera.
Data are reported as mean±SD. Statistical analysis was performed with the StatView 4.0 package (ABACUS Concepts). Differences were analyzed by unpaired 2-tailed Student t tests or by Welch t tests, as appropriate. Values of P<0.05 and P<0.01 were taken as significant.
Effects of Long-Term IGF-1 Treatment on TNF-α–Induced Adhesion Molecule Expression and Monocyte Adhesion in Endothelial Cells
First, we determined the effect of long-term IGF-1 treatment on TNF-α–induced ICAM-1, VCAM-1, and E-selectin expression. Because of the species specificity of the antibodies for adhesion molecules, we used HUVECs to determine the expression of ICAM-1, VCAM-1, and E-selectin (Figure 1). HUVECs were exposed to 10 ng/mL IGF-1 or vehicle for 16 hours. Cells were then stimulated with vehicle or TNF-α (5 ng/mL) for 6 hours. As shown in Figure 1, IGF-1 treatment (10 ng/mL for 22 hours) did not significantly increase ICAM-1, VCAM-1, or E-selectin expression. TNF-α treatment for 6 hours alone stimulated ICAM-1, VCAM-1, and E-selectin expression significantly (absorbance at 490-nm optical density was 0.37±0.008, 0.41±0.014, and 0.38±0.01 [mean±SD], respectively). Compared with 6 hours of TNF-α treatment alone, pretreatment with IGF-1 significantly enhanced TNF-α–mediated expression of ICAM-1, VCAM-1, and E-selectin expression (Figures 1A, 1B, and 1C).
To determine the biological significance of IGF-1 on TNF-α–induced adhesion molecule expression, we examined the ability of IGF-1 pretreatment to augment U937 cell adhesion to HUVECs. As shown in Figure 1D, treatment of HUVECs with 5 ng/mL TNF-α for 6 hours increased U937 cell adhesion to HUVECs (Figures 1E and 1F), whereas IGF-1 alone had no effect (Figure 1G). When HUVECs were pretreated with IGF-1 for 16 hours and then incubated with TNF-α for 6 hours, adhesion of U937 cells to HUVECs was significantly increased compared with TNF-α treatment alone for 6 hours (≈77% increase) (Figure 1H). In summary, these results demonstrate that long-term IGF-1 treatment enhances TNF-α–mediated adhesion molecule expression and monocyte adhesion to endothelial cells.
Effect of Long-Term IGF-1 Treatment on TNF-α–Induced c-Jun and NF-κB Transcriptional Activation in Endothelial Cells
It has been reported that c-Jun– and NF-κB–dependent events regulate adhesion molecule expression in endothelial cells.19 Therefore, we examined the effects of long-term IGF-1 treatment on c-Jun and NF-κB activation induced by TNF-α in endothelial cells. We transiently transfected BAECs with c-Jun and NF-κB reporter genes. After 24 hours of transfection, BAECs were exposed to 10 ng/mL IGF-1 or vehicle for 16 hours. Cells were then stimulated with vehicle or TNF-α (5 ng/mL) for 12 hours. As shown in Figures 2A and 2B (lanes 1 and 3), treatment with 5 ng/mL TNF-α led to a 2.8±0.1- and 2.0±0.2-fold stimulation of c-Jun and NF-κB activation, respectively, measured by reporter gene assay. Pretreatment with 10 ng/mL IGF-1 for 16 hours itself did not increase c-Jun and NF-κB activation (lane 2) but significantly enhanced TNF-α–induced c-Jun and NF-κB activation (3.5±0.2- and 2.6±0.1-fold increase, respectively; Figures 2A and 2B, lane 4). We also found a similar enhancing effect of IGF-1 on TNF-α–induced c-Jun and NF-κB activation in HUVECs (data not shown). These data suggest that long-term IGF-1 treatment potentiates c-Jun and NF-κB activation by TNF-α.
Effects of Long-Term IGF-1 Treatment on the Expression of c-Jun and NF-κB Signaling Molecules in Endothelial Cells
To assess whether IGF-1–mediated enhancement of c-Jun and NF-κB activity is due to increased expression of c-Jun and NF-κB signaling molecules, we measured the expression levels of MEKK1, MEKK3, MEK4, and JNK1/2 by Western blotting in BAECs. As shown in Figure 3A, there were no significant changes in protein expression levels. It is well known that several adapter proteins, including Gab1 (one of the members of the IRS-1 family) and 14-3-3β, are able to regulate mitogen-activated protein kinases (MAPKs).13,20⇓ Long-term IGF-1 treatment caused no significant differences in 14-3-3β protein expression assayed by Western blot. However, we found that 50 ng/mL IGF-1 significantly decreased the expression of both IRS-1 and Gab1 expression to ≈10% of control after 16 hours of treatment (Figures 3B and 3D). Decreases in Gab1 expression by IGF-1 treatment were concentration dependent, with an IC50 value of ≈10 ng/mL (Figure 3C). We also found a similar inhibitory effect of IGF-1 on Gab1 expression in HUVECs (data not shown). In contrast to Gab1 and IRS-1, we did not find any significant differences in IRS-2 expression (Figure 3B). These data suggest a role for Gab1 and IRS-1 in regulating c-Jun and NF-κB activation by TNF-α after long-term IGF-1 treatment.
Effects of Gab1 and IRS-1 Expression on Activation of c-Jun and NF-κB in Response to TNF-α in Endothelial Cells
We first investigated whether overexpression of Gab1 inhibited TNF-α–mediated c-Jun or NF-κB activation in BAECs, which have a higher transfection efficiency than HUVECs. BAECs were transiently transfected with pBTM-IRS-1, HA-tagged Gab1 (HA-Gab1), or vector alone with c-Jun or NF-κB reporter genes. TNF-α increased c-Jun activation, as shown in Figures 4A and 4C. Overexpression of Gab1 but not IRS-1 inhibited TNF-α–induced c-Jun activation. Interestingly, overexpression of both IRS-1 and Gab1 inhibited NF-κB transcriptional activity in a dose-dependent manner (Figures 4B and 4D). We also found a similar inhibitory effect of Gab1 on TNF-α–induced c-Jun and NF-κB activation in HUVECs (data not shown).
Next, to determine the role of Gab1 on the enhancing effect of IGF-1 on TNF-α–induced signaling, we evaluated whether overexpression of Gab1 can reverse the enhancing effect of IGF-1 on TNF-α–induced c-Jun and NF-κB activation. As shown in Figure 2, 10 ng/mL IGF-1 enhanced TNF-α (5 ng/mL)–induced c-Jun and NF-κB activation (compare lanes 3 and 4). We found that overexpression of Gab1 reversed IGF-1–mediated enhancement of c-Jun and NF-κB activation (compare lanes 4 and 6). At the highest concentration of Gab1, there was inhibition of TNF-α signaling (compare lanes 3 and 7). HA-Gab1 expression was dose dependent on cDNA amount (data not shown).
Gab1 Inhibits MEKK3-Mediated, but Not MEKK1- and MEK4-Mediated, c-Jun and NF-κB Transcriptional Activity
MEKK1 is MAPK kinase kinase-1, which has been reported to regulate ERK1/2, JNK, and NF-κB activation in several cell lines.21 Therefore, we evaluated the effect of Gab1 on c-Jun and NF-κB transcriptional activity induced by a CA form of MEKK1 (CA-MEKK1). As shown in Figure 5A, transient transfection of BAECs with CA-MEKK1 induced c-Jun and NF-κB transcriptional activity by 3- to 4-fold (Figures 5A and 5D). Overexpression of Gab1 did not significantly inhibit CA-MEKK1–stimulated c-Jun and NF-κB transcriptional activity (Figures 5A and 5D). MEKK3 is a MAPK kinase kinase specific for MKK4, MKK7, and MKK6, and it regulates JNK, p38, and NF-κB activation separately from the MEKK1 pathway.22 Recently, Yang et al23 reported that MEKK3 plays a critical role in TNF-α–induced NF-κB activation and directly phosphorylates IκB kinase (IKK). To examine the role of Gab1 in MEKK3-mediated c-Jun and NF-κB transcriptional activity, we transiently transfected CA-MEKK3 and determined c-Jun and NF-κB transcriptional activity by reporter gene assay. As shown in Figures 5B and 5E, CA-MEKK3 also induced a 3- to 4-fold increase in c-Jun and NF-κB transcriptional activity. Overexpression of Gab1 significantly inhibited CA-MEKK3–induced c-Jun and NF-κB transcriptional activity in a dose-dependent manner (Figures 5B and 5E). As shown in Figures 5C and 5F, a DN form of MEKK3 (DN-MEKK3) significantly inhibited TNF-α–induced c-Jun and NF-κB transcriptional activity, suggesting that MEKK3 is required for c-Jun and NF-κB transcriptional activity induced by TNF-α in endothelial cells and that the inhibitory effect of Gab1 on TNF-α–induced c-Jun and NF-κB transcriptional activity is due, at least in part, to inhibition of the MEKK3 pathway.
Because it is impossible to measure MEKK3 activity directly at present (probably because of the loss of an essential cofactor during immunoprecipitation as previously reported22), we determined the effect of Gab1 on CA-MEKK3–stimulated MEK4 and IKK activation. As shown in Figures 6A and 6B, we found that activation of both MEK4 and IKK induced by MEKK3 was attenuated by Gab1. In contrast, Gab1 had no effect on c-Jun activation induced by CA-MEK4 (Figure 6C). These data indicate that Gab1 specifically inhibits c-Jun transcriptional activity induced by MEKK3 (but not by MEKK1 and MEK4) and that the target of this inhibitory effect is downstream from MEKK3 but upstream from MEK4.
Gab1 Associates With MEKK3 In Vivo and In Vitro
Because it has been reported that Gab1 associates with ERK1/2, 24 we examined whether Gab1 can associate with MEKK3. We coexpressed Xpress-tagged Gab1 (Xp-Gab1) and Flag-tagged MEKK3 (Flag-MEKK3) in BAECs and performed coimmunoprecipitation assays. Immunoprecipitation of Flag-MEKK3 with the anti-Xpress antibody brought down Xp-Gab1 (Figure 6D). These data suggest that Gab1 forms a complex with MEKK3 in intact cells (Figure 6D).
The nature of the interaction between Gab1 and MEKK3 was further studied by coimmunoprecipitation in vitro with the use of well-characterized proteins (Figures 6E and 6F). As shown in Figure 6F, His-tagged Gab1 wild type (lanes 1 through 4) and either CA-MEKK3 (lane 2) or DN-MEKK3 (lanes 3 and 4) were expressed in vitro by using the rabbit reticulocyte lysate system and coimmunoprecipitated by using anti-His mouse monoclonal IgG2a (lanes 1 through 3) or anti-Flag mouse monoclonal IgG2a antibody as a control (lane 4) (Figure 6F). The observation that wild-type Gab1 could be coimmunoprecipitated with CA-MEKK3 or DN-MEKK3 but not with control anti-Flag antibody suggests that Gab1 is directly associated with MEKK3 and that this association is independent of MEKK3 kinase activity.
Expression of Gab1 Regulates CA-MEKK3–Induced ICAM-1, VCAM-1, and E-Selectin Expression in HUVECs
Both c-Jun and NF-κB are important mediators of adhesion molecule gene expression in HUVECs.19,25⇓ Because we found that Gab1 plays a critical role in TNF-α–induced and MEKK3-induced c-Jun and NF-κB transcriptional activity, we determined the effects of overexpression of Gab1 on CA-MEKK3–induced ICAM-1, VCAM-1, and E-selectin expression in HUVECs. As shown in Figures 7A through 7C, Gab1 significantly inhibited CA-MEKK3–induced ICAM-1, VCAM-1, and E-selectin expression (measured by ELISA) in a dose-dependent manner. These data suggest that Gab1 inhibits c-Jun and NF-κB transcriptional activity and decreases MEKK3-induced adhesion molecule expression.
Inflammatory stimuli, including TNF-α, increase adhesion molecule expression and contribute to the initial phase of atherogenesis.1 In the present study, the increased response to inflammatory stimuli demonstrated with IGF-1 treatment can be extrapolated to states of hyperinsulinemia and obesity.3,26,27⇓⇓ Specifically, reduction of Gab1 expression by long-term IGF-1 treatment in endothelial cells may contribute to the pathogenesis of atherosclerosis in hyperinsulinemia and obesity by promoting inflammation. The mechanism by which IGF-1 reduces IRS-1 family member expression is unclear. However, deVente et al28 have reported that protein kinase C decreases IRS-1 expression via a transcriptional mechanism. These data plus our observation of Gab1-induced and IRS-1–induced decreases after IGF-1 treatment suggest that this is an important regulatory pathway.
It has been reported that IGF-1 is growth promoting for VSMCs and plays a pathological role in cardiovascular diseases.4 Until now, the primary role of IGF-1 in endothelial cells has been to stimulate angiogenesis and tube formation.3,4⇓ In the present study, we found that a long-term low dose of IGF-1 (10 ng/mL) significantly enhances TNF-α–induced adhesion molecule expression in endothelial cells. These novel findings suggest a role for IGF-1 during the initial phase of atherosclerosis, related to vascular inflammation. Of interest, Balaram and colleagues29,30⇓ have reported that a high concentration of IGF-1 (150 ng/mL) increases ICAM-1 expression but that the effect of IGF-1 on ICAM-1 expression is much weaker than that of TNF-α.
There is increasing evidence to indicate that IGF-1 may have a critical role in plaque stability via regulating VSMC survival. Patel et al31 have reported that human plaque–derived VSMCs show an intrinsic sensitivity to apoptosis, which is caused in part by defective expression of the IGF-1 receptor, and impaired IGF-1–mediated survival signaling. Interestingly, Anwar et al5 have reported that TNF-α reduces IGF-1 in VSMCs, and this downregulation by TNF-α may play a role in acute coronary syndrome by decreasing VSMC viability and promoting plaque instability. On the basis of our findings, we propose that IGF-1 is an enhancing factor for cytokine-induced endothelial cell inflammation and may play a role in the early phase of atherogenesis. Further investigation will be needed to clarify the role of IGF-1 in early atherosclerosis formation.
In the present study, we found that Gab1 associates directly with MEKK3, and that Gab1 then inhibits MEKK3-mediated c-Jun and NF-κB transcription activity and adhesion molecule expression (Figure 8). Members of the mammalian MEKK family of serine/threonine kinases have been demonstrated to stimulate MEKK4 and JNK signaling.22 Of note, it has been reported that MEKK3 plays a major role in TNF-α–induced NF-κB activation and that MEKK3 directly phosphorylates IKK.23 Because MEKK3 is able to regulate both c-Jun and NF-κB pathways, MEKK3 activity may have a greater effect on the expression of adhesion molecules than other MAPK kinase kinases, such as ASK1 or TAK1. Recently, Roshan et al24 and Gual et al32 have reported that Gab1 associates directly with phosphorylated ERK2 and protein kinase C and that Gab1 acts as a substrate for these kinases. The mechanism of interaction between Gab1 and these serine/threonine kinases (including the role of serine/threonine phosphorylation of Gab1) remains unclear. We also determined whether Gab1 can coimmunoprecipitate with MEKK1, but we could not detect any association of Gab1 with MEKK1 (data not shown). However, we could not conclude that there is no association of Gab1 with MEKK1, because the detection of interaction by using coimmunoprecipitation is heavily dependent on the antibodies used.
In contrast to our results, Garcia-Guzman et al20 have reported that the complex formation of tyrosine-phosphorylated Gab1 and Crk is correlated with stimulation of the JNK pathway. One explanation for this difference is that Garcia-Guzman studied hepatocyte growth factor, which stimulated Gab1 tyrosine phosphorylation by ≈5- to 10-fold,20 whereas tyrosine phosphorylation of Gab1 by TNF-α was increased only ≈1.5-fold (data not shown). These results suggest that Gab1-MEKK3 interaction and MEKK3 activation may be regulated by Gab1 phosphorylation, an important area for future investigation.
This work was supported by grants from the NIH to Dr Abe (HL-61319) and to Dr Berk (HL-44721 and HL-49192).
↵*These authors contributed equally to this study.
Original received November 27, 2001; revision received April 26, 2002; accepted April 26, 2002.
- ↵Bayes-Genis A, Conover CA, Schwartz RS. The insulin-like growth factor axis: a review of atherosclerosis and restenosis. Circ Res. 2000; 86: 125–130.
- ↵Anwar A, Zahid AA, Scheidegger KJ, Brink M, Delafontaine P. Tumor necrosis factor-α regulates insulin-like growth factor-1 and insulin-like growth factor binding protein-3 expression in vascular smooth muscle. Circulation. 2002; 105: 1220–1225.
- ↵Yan C, Takahashi M, Okuda M, Lee JD, Berk BC. Fluid shear stress stimulates big mitogen-activated protein kinase 1 (BMK1) activity in endothelial cells: dependence on tyrosine kinases and intracellular calcium. J Biol Chem. 1999; 274: 143–150.
- ↵Surapisitchat J, Hoefen RJ, Pi X, Yoshizumi M, Yan C, Berk BC. Fluid shear stress inhibits TNF-α activation of JNK but not ERK1/2 or p38 in human umbilical vein endothelial cells: inhibitory crosstalk among MAPK family members. Proc Natl Acad Sci U S A. 2001; 98: 6476–6481.
- ↵Basta G, Lazzerini G, Massaro M, Simoncini T, Tanganelli P, Fu C, Kislinger T, Stern DM, Schmidt AM, De Caterina R. Advanced glycation end products activate endothelium through signal-transduction receptor RAGE: a mechanism for amplification of inflammatory responses. Circulation. 2002; 105: 816–822.
- ↵Takahashi-Tezuka M, Yoshida Y, Fukada T, Ohtani T, Yamanaka Y, Nishida K, Nakajima K, Hibi M, Hirano T. Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to ERK mitogen-activated protein kinase. Mol Cell Biol. 1998; 18: 4109–4117.
- ↵Guan Z, Buckman SY, Pentland AP, Templeton DJ, Morrison AR. Induction of cyclooxygenase-2 by the activated MEKK1→SEK1/MKK4→p38 mitogen-activated protein kinase pathway. J Biol Chem. 1998; 273: 12901–12908.
- ↵Chao TH, Hayashi M, Tapping RI, Kato Y, Lee JD. MEKK3 directly regulates MEK5 activity as part of the big mitogen-activated protein kinase 1 (BMK1) signaling pathway. J Biol Chem. 1999; 274: 36035–36038.
- ↵Yoshizumi M, Abe J, Haendeler J, Huang Q, Berk BC. Src and cas mediate JNK activation but not ERK1/2 and p38 kinases by reactive oxygen species. J Biol Chem. 2000; 275: 11706–11712.
- ↵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.
- ↵Yan C, Luo H, Lee JD, Abe J, Berk BC. Molecular cloning of mouse ERK5/BMK1 splice variants and characterization of ERK5 functional domains. J Biol Chem. 2001; 276: 10870–10878.
- ↵Heinlein CA, Ting HJ, Yeh S, Chang C. Identification of ARA70 as a ligand-enhanced coactivator for the peroxisome proliferator-activated receptor γ. J Biol Chem. 1999; 274: 16147–16152.
- ↵Luscinskas FW, Kansas GS, Ding H, Pizcueta P, Schleiffenbaum BE, Tedder TF, Gimbrone MA Jr. Monocyte rolling, arrest and spreading on IL-4-activated vascular endothelium under flow is mediated via sequential action of L-selectin, β1-integrins, and β2-integrins. J Cell Biol. 1994; 125: 1417–1427.
- ↵Min W, Pober JS. TNF initiates E-selectin transcription in human endothelial cells through parallel TRAF-NF-κB and TRAF-RAC/CDC42-JNK-c-Jun/ATF2 pathways. J Immunol. 1997; 159: 3508–3518.
- ↵Deacon K, Blank JL. Characterization of the mitogen-activated protein kinase kinase 4 (MKK4)/c-Jun NH2-terminal kinase 1 and MKK3/p38 pathways regulated by MEK kinases 2 and 3: MEK kinase 3 activates MKK3 but does not cause activation of p38 kinase in vivo. J Biol Chem. 1997; 272: 14489–14496.
- ↵Roshan B, Kjelsberg C, Spokes K, Eldred A, Crovello CS, Cantley LG. Activated ERK2 interacts with and phosphorylates the docking protein GAB1. J Biol Chem. 1999; 274: 36362–36368.
- ↵Sowers JR, Epstein M. Diabetes mellitus and associated hypertension, vascular disease, and nephropathy: an update. Hypertension. 1995; 26: 869–879.
- ↵deVente JE, Carey JO, Bryant WO, Pettit GJ, Ways DK. Transcriptional regulation of insulin receptor substrate 1 by protein kinase C. J Biol Chem. 1996; 271: 32276–32280.
- ↵Patel VA, Zhang QJ, Siddle K, Soos MA, Goddard M, Weissberg PL, Bennett MR. Defect in insulin-like growth factor-1 survival mechanism in atherosclerotic plaque-derived vascular smooth muscle cells is mediated by reduced surface binding and signaling. Circ Res. 2001; 88: 895–902.