|Home | About | Journals | Submit | Contact Us | Français|
We investigated the role of proline-rich tyrosine kinase 2 (Pyk2) in the mechanism of NF-κB activation and endothelial cell (EC) inflammation induced by thrombin, a procoagulant serine protease released in high amounts during sepsis and other inflammatory conditions. Stimulation of ECs with thrombin resulted in a time-dependent activation of Pyk2. RNA interference knockdown of Pyk2 attenuated thrombin-induced activity of NF-κB and expression of its target genes, vascular cell adhesion molecule-1 and monocyte chemoattractant protein-1. Pyk2 knockdown impaired thrombin-induced activation of IκB kinase (IKK) and phosphorylation (Ser32 and Ser36) of IkappaBα, but, surprisingly, failed to prevent IκBα degradation. However, depletion of IKKα or IKKβ was effective in inhibiting IκBα phosphorylation/degradation, as expected. Intriguingly, Pyk2 knockdown impaired nuclear translocation and DNA binding of RelA/p65, despite the inability to prevent IκBα degradation. In addition, Pyk2 knockdown was associated with inhibition of RelA/p65 phosphorylation at Ser536, which is important for transcriptional activity of RelA/p65. Depletion of IKKα or IKKβ each impaired RelA/p65 phosphorylation. Taken together, these data identify Pyk2 as a critical regulator of EC inflammation by virtue of engaging IKK to promote the release and the transcriptional capacity of RelA/p65, and, additionally, by its ability to facilitate the nuclear translocation of the released RelA/p65. Thus, specific targeting of Pyk2 may be an effective anti-inflammatory strategy in vascular diseases associated with EC inflammation and intravascular coagulation.
Endothelial cell (EC) inflammation associated with intravascular coagulation is an important component of many diseases, including acute lung injury and acute respiratory distress syndrome. However, the mechanisms by which the procoagulant thrombin promotes EC inflammation are not fully understood. This study identifies proline-rich tyrosine kinase 2 as a critical determinant of EC inflammation by virtue of promoting the release, increasing the transcriptional capacity, and facilitating the nuclear translocation of the released RelA/p65.
The NF-κB controls myriad biological effects, including immune and inflammatory responses, cell fate decisions, such as proliferation, differentiation, tumorigenesis, and apoptosis, and is implicated in many disease states, such as atherosclerosis, acute lung injury, arthritis, inflammatory bowel diseases, and chronic obstructive pulmonary diseases (1–8). The mammalian NF-κB family consists of five members: RelA (p65), RelB, c-Rel, p50, and p52 (1–3). Of these, RelA/p65, c-Rel, and RelB are transcriptionally active, due to the presence of both transactivation and DNA-binding domains, whereas p50 and p52 serve primarily as DNA-binding subunits, as they lack the transactivation domain (1, 2). NF-κB, typically a heterodimer of p50 and RelA/p65 subunits, is sequestered in an inactive state in the cytoplasm through its association with IκBα that masks the nuclear localization signal of RelA/p65 (2). Activation of NF-κB requires phosphorylation of IκBα at Ser32 and Ser36 by a macromolecular IκB kinase (IKK) complex containing the catalytic subunits, IKKα and IKKβ, and the regulatory subunit, NF-κB essential modulator (NEMO)/IKKγ (1, 2). Phosphorylation initiates rapid polyubiquitination of IκBα by the E3 Skp1-cullin-F-box protein β-transducin repeat-containing protein (E3-SCFβ-TrCP) ubiquitin ligase, targeting it for degradation by the 26 S proteasome (1, 2). The released NF-κB undergoes nuclear translocation, binds to cognate κB enhancer elements in the target genes, and activates their transcription (1–3, 9, 10). Another component of NF-κB activation involves phosphorylation of RelA/p65 that serves to increase the transcriptional capacity of the bound NF-κB. Studies have shown that phosphorylation at serine 276, 311, 529, or 536 increases the transactivation potential of RelA/p65 in a cell- and stimulus-specific manner (1–3, 11).
Among the key targets of NF-κB in endothelial cells (ECs) are vascular cell adhesion molecule (VCAM)-1 and monocyte chemoattractant protein (MCP)-1/CC-chemokine ligand 2. VCAM-1 is an inducible adhesive protein that interacts with its counter receptor, very late antigen-4 (CD49 d/CD29) present on the circulating monocytes and lymphocytes (12). MCP-1 is a secreted protein that specifically attracts monocytes to the site of inflammation (13). The coordinate action of VCAM-1 and MCP-1 serves to promote the adhesion of monocytes to the vascular endothelium and, subsequently, their transendothelial migration (TEM) to the peripheral tissues, resulting in tissue injury (12–14). Activation of NF-κB and, thereby, expression of VCAM-1 and MCP-1 on the vascular endothelium is induced by a variety of proinflammatory mediators, including the procoagulant thrombin, a serine protease activated during intravascular coagulation initiated by tissue injury or sepsis (15). We and others have shown that thrombin-induced NF-κB–dependent gene expression is initiated by ligation of the protease-activated receptor-1 and activation of downstream signals (9, 16, 17). A key signal involves activation of the protein kinase C-δ, which regulates NF-κB activation by inducing IKK-dependent release, and p38 mitogen-activated protein kinase (MAPK)–dependent phosphorylation of RelA/p65 (10, 17).
Proline-rich tyrosine kinase 2 (Pyk2), also known as calcium-dependent tyrosine kinase, cell-associated tyrosine kinase-β, and related adhesion focal tyrosine kinase, is a nonreceptor tyrosine kinase and a member of the focal adhesion kinase family. It is expressed in various cell types, including neuronal cells and fibroblasts, but is abundant in cells of hematopoietic origin, especially B lymphocytes and monocytes/macrophages (18–20). Pyk2 is also present in different types of ECs, and has been implicated in various signaling pathways (21, 22). It is an important regulator of endothelial permeability and angiogenesis (23, 24), and has been shown to promote TEM of leukocytes via intercellular adhesion molecule-1 (ICAM-1)–dependent tyrosine phosphorylation of vascular endothelial cadherins and expression of IL-8 and MCP-1 (25–27). Thrombin has been shown to activate Pyk2 in ECs in a Ca2+-dependent manner (21); however, the role of Pyk2 in thrombin-induced NF-κB activation and proinflammatory gene expression remains unclear. In this study, we ascertained the role of Pyk2 in the mechanism of these responses. Our data show a pivotal role of Pyk2 in causing EC inflammation by its ability to activate IKK to promote the release and the transcriptional capacity of RelA/p65, and also by virtue of facilitating the nuclear translocation of RelA/p65.
Human thrombin was purchased from Enzyme Research Laboratories (South Bend, IN). Tyrphostin A9 was purchased from Calbiochem-Novabiochem Corp. (La Jolla, CA). SB 203,580 was purchased from Enzo Life Sciences (Farmingdale, NY). Polyclonal antibodies to VCAM-1, RelA/p65, β-actin, IκBα, Cu-Zn superoxide dismutase-1, p38 MAPK and RNA polymerase II were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies to phospho-(Ser180/Ser181)-IKKα/IKKβ, phospho-(Ser32 and Ser36)-IκBα, phospho-(Thr180/Tyr182)-p38 MAPK, phospho-(Ser536)-RelA/p65, phospho-(Tyr402)-Pyk2, IKKα, IKKβ, Pyk2, and casein kinase (CK) II were obtained from Cell Signaling (Beverly, MA). All other materials were from VWR Scientific Products Corp. (Gaithersburg, MD).
Human pulmonary artery ECs (HPAECs) were purchased from Lonza (Walkersville, MD). HPAECs were cultured, as previously described (28), in gelatin-coated flasks using endothelial basal medium 2 with bullet kit additives (BioWhittaker, Walkersville, MD). HPAECs used in the experiments were between three and six passages.
After treatment, cell lysates were prepared and analyzed by immunoblotting, as previously described (29). For immunoprecipitation, cell lysates were precleared with 10 μl of protein A/G-agarose beads for 4 hours at 4°C. The precleared lysates were immunoprecipitated by incubating with 0.6–1.0 μg of appropriate antibody and 10 μl of the protein A/G-agarose beads at 4°C overnight with gentle shaking, as previously described (29). The immunoprecipitates were washed four times with the same volume of ice-cold phosphorylation lysis buffer and then extracted by boiling with 2× SDS sample buffer for 5 minutes. The extracted proteins were then analyzed by immunoblotting (29).
Total cell lysates were analyzed by ELISA for measurement of MCP-1 using the human cardiovascular disease panel Multiplex kit (Millipore, Bradford, MA), according to manufacturer’s recommendations.
SMARTpool short interfering RNA duplexes targeting Pyk2, IKKα, IKKβ, or CKII, and a nonspecific siRNA control were obtained from Dharmacon (Lafayette, CO). HPAECs were transfected with appropriate siRNAs using DharmaFect1 siRNA Transfection Reagent (Dharmacon), as previously described (28). Briefly, 50–100 nM siRNA was mixed with DharmaFect1 and added to cells that were approximately 50–60% confluent. After 24–36 hours, cell lysates were prepared and subjected to immunoblotting to assess the levels of Pyk2, CKII, IKKα, or IKKβ.
The plasmid, pNF-κB-luciferase, containing five copies of consensus NF-κB sequences linked to a minimal E1B promoter-luciferase gene, was from Stratagene (La Jolla, CA). The constructs, Gal4-p65 (286–551), Gal4-p65 (286–551, S536A), Gal4 Vector, and Gal4-luciferase (LUC), were kindly provided by M. W. Mayo (University of Virginia, Charlottesville, VA). Transfections were performed using the diethylaminoethyl (DEAE)-dextran method, essentially as previously described (29). A transfection efficiency of 16 (±3)% (mean [±SD]; n = 3) was achieved with this protocol. The plasmid, pTKRLUC (Promega, Madison, WI), containing Renilla luciferase reporter gene driven by the constitutively active thymidine kinase promoter, was used to normalize the transfection efficiency. Cell extracts were assayed for Firefly and Renilla luciferase activities using the Promega Biotech Dual Luciferase Reporter Assay System. The data were expressed as a ratio of Firefly to Renilla luciferase activity.
After treatment, cytoplasmic and nuclear extracts were prepared essentially as described previously (28). The purity of nuclear and cytoplasmic extracts was confirmed by the presence or absence of RNA polymerase II in these extracts (data not shown). The concentration of the cytoplasmic and nuclear protein was determined using a Bio-Rad protein determination kit (Bio-Rad, Hercules, CA).
Electrophoretic mobility shift assay was performed as previously described (29). Briefly, 10 μg of nuclear extract was mixed with 1 μg of poly (dI-dC) in a binding buffer (10 mM Tris-HCl [pH 7.5], 50 mM NaCl, 0.5 mM DTT, 10% glycerol [20 μl final volume]) for 15 minutes at room temperature. The reaction mixture was incubated with [γ-32P]-labeled double-stranded oligonucleotides containing NF-κB consensus sequence (5-GTTGAGGGGACT-TTCCCAGGC-3′) (30) for 15 minutes at room temperature. The DNA–protein complexes were subjected to electrophoresis on a 5% native gel, dried on Whatman paper (Bio-Rad), and then exposed to radioactivity.
Data are expressed as means (±SE). Statistical significance was determined using one-way ANOVA followed by Tukey’s post hoc test for multigroup comparisons using Prism 5.0 software (GraphPad Software, San Diego, CA). In some cases (Figures 1A and 1B), Student’s t test was performed for comparisons between experimental groups. A P value less than 0.05 was considered significant.
To determine the involvement of Pyk2 in the mechanism of NF-κB activation and proinflammatory gene expression, we first determined the time course of Pyk2 activation by thrombin in ECs. Because activation of Pyk2 is initiated through its phosphorylation at Tyr402 (31), we determined this response as an indication of Pyk2 activation. Thrombin induced Tyr402 phosphorylation of Pyk2 in a biphasic manner, with first peak phosphorylation occurring between 1 and 5 minutes and the second peak phosphorylation between 0.5 and 2 hours (Figure 1A and data not shown). We next examined whether Pyk2 is required for thrombin-induced NF-κB activity. To assess this possibility, we evaluated the effect of RNA interference knockdown of Pyk2 on NF-κB activity. Cells were transfected with siRNA targeting Pyk2 (si-Pyk2) or control siRNA, and Pyk2 expression was analyzed by immunoblotting. A marked depletion of Pyk2 was observed in si-Pyk2–transfected cells compared with the cells transfected with control siRNA (Figure 1B). Importantly, si-Pyk2 had no effect on the levels of focal adhesion kinase (Figure 1B), a structurally related family member with approximately 48% amino acid identity (20), indicating the specificity of Pyk2 knockdown by this approach. Depletion of Pyk2 impaired thrombin-induced NF-κB activity (Figure 1C). We also determined whether Pyk2 influences NF-κB activity in response to TNF-α, a prototype of NF-κB activator. Depletion of Pyk2 was effective in inhibiting TNF-α–induced NF-κB activity as well (Figure 1D).
The above data led us to examine whether Pyk2 knockdown produces a similar effect on VCAM-1, a well defined NF-κB target gene in ECs (3, 16, 17). Depletion of Pyk2 inhibited thrombin-induced VCAM-1 expression (Figure 2A), consistent with its effect on NF-κB activity (Figure 1C). To determine whether kinase activity of Pyk2 is required, we assessed the effect of inhibiting Pyk2 on this response. Pretreatment of cells with tyrphostin A9, an inhibitor of Pyk2, abolished thrombin-induced VCAM-1 expression (Figure 2B). We also assessed the effect of depleting Pyk2 on MCP-1, another well defined NF-κB target gene (32). Results showed that Pyk2 knockdown was effective in suppressing the basal as well as thrombin-induced expression of MCP-1 (Figure 2C).
Phosphorylation of IκBα at Ser32 and Ser36, and its subsequent degradation, is a requirement for the release of RelA/p65 for its translocation to the nucleus. Because phosphorylation of IκBα is mediated by IKK complex, we first evaluated the role of Pyk2 in IKK activation. Depletion of Pyk2 inhibited activation of IKK upon thrombin challenge, as determined by phosphorylation of IKKβ at Ser181 (Figure 3A). We next examined whether knockdown of Pyk2 affects phosphorylation and degradation of IκBα. Previously, our time course experiments established that thrombin-induced IκBα phosphorylation/degradation peaks at between 1 and 2 hours after stimulation (28, 33). We therefore used this time point in subsequent experiments to address the role of Pyk2 in the mechanism of NF-κB activation. Depleting Pyk2 was effective in inhibiting IκBα phosphorylation, but, surprisingly, had no effect on IκBα degradation (Figures 3B and 3C). However, depletion of IKKα or IKKβ both reduced phosphorylation as well as degradation of IκBα, as expected (Figure 3D). These results raise the possibility that Pyk2 knockdown exerts two effects on IκBα degradation: one causes inhibition of IKK-dependent and the other promotes IKK-independent degradation of IκBα, resulting in no net inhibition of thrombin-induced IκBα degradation in Pyk2-depeleted cells.
We next investigated the effect of depleting Pyk2 on nuclear translocation and DNA binding of RelA/p65. Analyses of nuclear extracts from control and thrombin-challenged cells by immunoblotting and electrophoretic mobility shift assay showed a marked inhibition in nuclear translocation and DNA binding of RelA/p65 in cells transfected with si-Pyk2 (Figures 4A and 4B). Together, these results show impaired nuclear translocation and DNA binding of RelA/p65 despite the degradation of IκBα in cells depleted of Pyk2 (Figure 3C). Considered together, these results point to a role of Pyk2 in facilitating RelA/p65 translocation to the nucleus.
We also determined whether Pyk2 contributes to the transactivation potential of RelA/p65, in addition to promoting its nuclear translocation and DNA binding function. To this end, Pyk2-depleted cells, or cells pretreated with tyrphostin A9 were challenged with thrombin, and total cell lysates were analyzed for Ser536 phosphorylation of RelA/p65. Depletion or inhibition of Pyk2 each caused a significant inhibition in RelA/p65 phosphorylation induced by thrombin (Figures 5A and 5B). To definitively establish that Ser536 phosphorylation is critical for thrombin-induced transcriptional activity of RelA/p65, we used a construct Gal4-p65 that encodes a fusion protein containing DNA-binding domain of the yeast Gal4 transcription factor and transactivation domain (residues 286–551) of RelA/p65 (34). Cells were cotransfected with Gal4-p65 and Gal4-LUC, which contains four copies of Gal4 DNA-binding consensus sites from yeast GAL4 gene promoter (34), and then challenged with thrombin. Expression of nuclear fusion protein Gal4-p65 caused a marked increase in constitutive transactivation of Gal4-LUC (Figure 5C), as expected. A further increase in transactivation by Gal4-p65 was noted upon stimulation with thrombin (Figure 5C). However, both basal and thrombin-induced transactivation potential of Gal4-p65 was lost upon mutation of Ser536 in transactivation domain of RelA/p65 (Figure 5C). These results indicate an essential role of Ser536 in controlling thrombin-induced transcriptional activity of RelA/p65.
We previously demonstrated a role of p38 MAPK in regulating thrombin-induced Ser536 phosphorylation of RelA/p65 in ECs (35). We therefore asked whether Pyk2 is required for p38 MAPK activation. Depletion of Pyk2 failed to inhibit p38 MAPK activation, as determined by its phosphorylation at Thr180/Tyr182 (see Figure E1 in the online supplement). These data suggest the existence of an additional pathway for Ser536 phosphorylation of RelA/p65 that may be engaged by thrombin in a Pyk2-dependent manner. Given that Pyk2 depletion prevents activation of IKK (Figure 3A), and that IKK is implicated in Ser536 phosphorylation of RelA/p65 (36), we addressed the role of IKK in this response. We observed that depletion of IKKα and IKKβ each inhibited RelA/p65 phosphorylation by thrombin (Figure 6). Unlike RelA/p65 phosphorylation, depletion of IKKα or IKKβ showed no effect on p38 MAPK phosphorylation (Figure E2A). We also examined the other possibility, that p38 MAPK may regulate IKK activation, by monitoring the effect of p38 MAPK inhibition on IκBα phosphorylation on Ser32 and Ser36. Inhibiting p38 MAPK failed to prevent IκBα phosphorylation induced by thrombin (Figure E2B). These data are in line with our earlier studies showing that thrombin-induced IκBα degradation and RelA/p65 nuclear DNA binding were resistant to p38 MAPK inhibition (10, 35). Taken together, these data are consistent with a model wherein thrombin engages Pyk2 as an additional mechanism of RelA/p65 phosphorylation that relies on IKK. To further verify this model, we evaluated the effect of Pyk2 knockdown and p38 MAPK inhibition on Ser536 phosphorylation of RelA/p65, and found that this approach resulted in additive inhibitory effect on thrombin-induced RelA/p65 phosphorylation (Figure E3).
The nonreceptor tyrosine kinase, Pyk2, is an important regulator of EC barrier function and angiogenesis (23, 24). Recently, it has also been implicated in proinflammatory gene expression and leukocyte TEM (26, 27); however, the mechanistic basis of these responses is only partially understood. Because NF-κB is an essential regulator of genes involved in TEM of leukocytes, the present study was undertaken to address the role of Pyk2 in the mechanism of NF-κB activation in ECs. Our data show that Pyk2 is an important determinant of thrombin-induced EC inflammation, and that it mediates this response by activating IKK to promote the release and the transcriptional capacity of RelA/p65, and additionally by facilitating the nuclear translocation of released RelA/p65. Importantly, Pyk2 also controls NF-κB activation induced by TNF-α, suggesting that Pyk2 is a general regulator of NF-κB signaling in the endothelium.
We began our investigation by defining the time course of Pyk2 activation by thrombin. We noted a biphasic phosphorylation of Pyk2: an early and transient phase occurring between 1 and 5 minutes, followed by a late phase that begins at 30 minutes and is sustained for up to 2 hours after thrombin stimulation. The early phase of Pyk2 activation is in accord with a previous report (21) showing a similar time course of Pyk2 phosphorylation; however, the late phase was not detected in the earlier study, as Pyk2 phosphorylation was followed only up to 10 minutes after thrombin challenge. Thus, our data reveal a second phase (in addition to the early phase) of Pyk2 phosphorylation, and are consistent with two-stage mobilization of Ca2+, a requirement for Pyk2 activation, induced by thrombin in ECs (37, 38). Pyk2 activation was also observed in LPS-stimulated human umbilical vein ECs and human dermal microvascular ECs (26, 27), as well as in cyclic strain-exposed bovine aortic ECs (22). Thus, Pyk2 appears to be responsive to the proinflammatory environment in ECs from different vascular beds.
The activation of Pyk2 led us to assess its involvement in NF-κB activation. To this end, we evaluated the effect of RNAi knockdown of Pyk2 on NF-κB–dependent reporter activity. Depletion of Pyk2 abolished thrombin-induced NF-κB activity, indicating the requirement of Pyk2 in the response. Our findings are consistent with those of earlier studies indicating the involvement of Pyk2 in G protein–coupled receptor (GPCR)-mediated signaling and NF-κB activation (18, 39). We also examined if the impaired NF-κB activity secondary to Pyk2 knockdown leads to reduced expression of VCAM-1 and MCP-1. A significant decrease in the expression of these genes was observed after thrombin stimulation of Pyk2-depleted cells. Notably, the role of Pyk2 in NF-κB–dependent gene expression is not restricted to thrombin and the endothelium, but can also be induced by a variety of stimuli and in different cell types. Consistent with this, we found that depletion of Pyk2 was effective in inhibiting TNF-α–induced NF-κB activity in ECs. Other studies showing a role of Pyk2 in LPS-induced expression of IL-8 and MCP-1 in ECs, carbacol-induced NF-κB–dependent reporter activity in HeLa cells, and LPS- and peptidoglycan-induced IL-1β expression in macrophages (26, 27, 39, 40) further underscore the importance of Pyk2 as a common mediator of NF-κB signaling.
Our study yields the mechanistic basis of NF-κB activation by Pyk2. We demonstrate that Pyk2 mediates NF-κB activation by promoting the release, increasing the transcriptional capacity, and facilitating the nuclear translocation and DNA binding of RelA/p65. Nuclear translocation and DNA binding of RelA/p65 is contingent upon its release secondary to Ser32 and Ser36 phosphorylation, and, subsequently, degradation of IκBα. Consistent with this, Pyk2 knockdown impaired the phosphorylation of IKK and IκBα. Similarly, depletion of IKKα or IKKβ also inhibited IκBα phosphorylation and degradation. Intriguingly, however, Pyk2 depletion had no effect on IκBα degradation by thrombin. This does not seem to be the result of a nonspecific effect of Pyk2-siRNA, as IκBα level remained intact in unstimulated cells (Figure 3C). One possible explanation could be that a Ser32 and Ser36 phosphorylation–independent (i.e., IKK independent) mechanism of IκBα degradation is activated by thrombin in the absence of Pyk2. Indeed, studies have shown that phosphorylation of IκBα-PEST (proline–glutamic acid–serine-threonine) sequences by CKII-dependent mechanism also leads to its degradation (41, 42). It is likely that the inhibitory effect of Pyk2 depletion on IKK-dependent IκBα degradation is masked by CKII-dependent degradation of IκBα that may be activated by thrombin in the absence of Pyk2. Such a possibility is further supported by the requirement of Ca2+ for the latter mode of IκBα degradation (41). Because Ca2+ is also required for Pyk2 activity, it is plausible that the depletion of Pyk2 increases the availability of Ca2+ to facilitate CKII-mediated PEST sequence phosphorylation–dependent degradation of IκBα by thrombin. In support of this model, we found that RNAi knockdown of CKII was effective in protecting IκBα degradation only in cells depleted of Pyk2 (Figure E4). However, additional studies are required to unequivocally address this possibility.
Interestingly, Pyk2 knockdown was associated with impaired nuclear translocation and DNA binding of RelA/p65 in the face of IκBα degradation. These results suggest that Pyk2 may also be involved in facilitating the translocation of released RelA/p65 to the nucleus. We recently found that dynamic changes in actin cytoskeleton induced by thrombin via RhoA/Rho-associated kinase (ROCK)/Cofilin pathway are necessary for nuclear translocation of RelA/p65 (43, 44). Given the involvement of Pyk2 and its downstream kinase c-Src in activating RhoA/ROCK pathway (19, 45, 46), it is possible that Pyk2 promotes nuclear translocation and, consequently, DNA binding of RelA/p65 by causing alterations in the actin cytoskeleton. Consistent with this possibility, inhibition of Pyk2 by tyrphostin A9 attenuates thrombin-induced Ser3 phosphorylation and, thereby, inactivation of cofilin1 (K.M.B., unpublished results), an actin-binding protein that occupies a central position in Rho–actin pathway mediating RelA/p65 nuclear translocation (43). However, in view of the reports that Pyk2 can also function downstream of RhoA/ROCK (47, 48), the other possibility that RhoA/ROCK engages Pyk2 to mediate the above responses cannot be excluded. Further studies are needed to establish how Pyk2 and RhoA/ROCK are linked in the mechanism of NF-κB activation by thrombin.
An important event implicated in enhancing the transactivation potential of RelA/p65 includes its phosphorylation on serine residues, including Ser536 (10). We previously showed that thrombin induces Ser536 phosphorylation of RelA/p65, and that this response is mediated, in part, by p38 MAPK (34, 49). Because Pyk2 is required for p38 MAPK activation (50), and Pyk2 depletion reduces Ser536 phosphorylation of RelA/p65, we reasoned that Pyk2 may act upstream of p38 MAPK to mediate this response (RelA/p65 phosphorylation). However, our data that thrombin activates p38 MAPK in a Pyk2-independent manner argues against the involvement of Pyk2/p38 MAPK in RelA/p65 phosphorylation, and suggests the existence of an additional pathway that Pyk2 uses for this response. Indeed, we found that Pyk2 engages IKK (IKKα and IKKβ) to promote RelA/p65 phosphorylation and, thereby, NF-κB signaling induced by thrombin. Our findings that inhibition of p38 MAPK fails to prevent thrombin-induced IκBα phosphorylation at Ser32 and Ser36 (Figure E2B), an IKK-mediated event required for IκBα degradation and subsequently nuclear DNA binding of RelA/p65 (10, 35), also argues against the engagement of p38 MAPK upstream of IKK to mediate RelA/p65 phosphorylation. Thus, the inability of Pyk2 to mediate thrombin-induced p38 MAPK activation compared with an important role of Pyk2/p38 MAPK signaling in LPS-induced IL-8 and MCP-1 expression and vascular endothelial growth factor–induced EC migration (26, 27, 50) reveals a stimulus-specific activation of p38 MAPK by Pyk2 in ECs, and shows a fundamental signaling difference between thrombin and other receptors (GPCR versus LPS and vascular endothelial growth factor). It should also be stressed that the dominant inhibitory effect of Pyk2 depletion on thrombin-induced NF-κB activation and EC inflammation may derive from the ability of Pyk2 to regulate RelA/p65 at multiple levels (release of RelA/p65 from IκBα, nuclear translocation, and the transcriptional capacity of the released RelA/p65). Unlike Pyk2, p38 MAPK contributes to NF-κB activity via RelA/p65 phosphorylation alone. Moreover, in a separate ongoing study, we have found that Ser536 phosphorylation is dependent upon IκBα degradation (K.M.B., unpublished results). Because p38 MAPK is not required for IκBα degradation (10), its contribution to RelA/p65 phosphorylation and, thereby, EC inflammation may rely on Pyk2-dependent IκBα degradation induced by thrombin.
In summary, this study identifies Pyk2 as a critical determinant of NF-κB activation and EC inflammation. We show that Pyk2 engages IKK to liberate RelA/p65 from IκBα and empower the liberated RelA/p65 with the transcriptional capacity. Additionally, Pyk2 is engaged to facilitate the nuclear accumulation and DNA binding of the released RelA/p65. Together, these events serve to promote EC inflammation. Thus, specific targeting of Pyk2 may be an effective anti-inflammatory strategy in vascular diseases associated with EC inflammation and intravascular coagulation.
This work was supported in part by National Heart, Lung, and Blood Institute grants HL67424 and HL096907, American Lung Association Biomedical Research grant RG-169581-N, and American Heart Association Scientist Development Grant SDG5440016.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1165/rcmb.2012-0047OC on July 27, 2012