Cytokine expression and AIF-1-mediated activation of Rac2 in vascular smooth muscle cells: a role for Rac2 in VSMC activation

Ying Tian, Michael V. Autieri

Abstract

Allograft inflammatory factor-1 (AIF-1) is a cytoplasmic, calcium-binding, inflammation-responsive scaffold protein involved in vascular smooth muscle cell (VSMC) migration and proliferation. The objective of this study is to characterize AIF-1 functional protein interactions that may regulate VSMC activation. Through use of a bacterial two-hybrid screen, we identified a molecular interaction between AIF-1 and the small GTPase, Rac2, which was verified by pull-down and colocalization experiments. This was unexpected in that Rac2 expression had been considered to be restricted to hematopoietic cells. The Rac2/AIF-1 interaction is functional, in that a loss-of-function, point-mutated AIF-1 does not interact with Rac2; Rac2 colocalizes with AIF-1 in the cytoplasm of VSMC and cotranslocates to lamellopodia upon platelet-derived growth factor stimulation; and AIF-1 expression in VSMC leads to Rac2 activation. Because Rac2 function in VSMC had not been described, we focused on characterization of its function in these cells. Rac2 protein expression in VSMC is inducible by inflammatory cytokines, and Rac2 activation in VSMC is also responsive to inflammatory cytokines. Rac2 expression and activation patterns differ from the ubiquitously expressed Rac1. We hypothesized that Rac2 participates in VSMC activation. Retroviral overexpression of Rac2 in primary VSMC leads to increased migration, activation of the NADPH oxidation cascade, and increased activation of the Rac2 effector protein Pak1 and its proximal effectors, ERK1/2, and p38 (P < 0.05 for all). The major points of this study indicate a functional interaction between AIF-1 and Rac2 in VSMC leading to Rac2 activation and a potential function for Rac2 in inflammation-driven VSMC response to injury.

  • allograft inflammatory factor-1
  • signal transduction

intimal hyperplasias subsequent to mechanical and immunological insult remain clinically significant obstacles limiting the success of vascular interventions and solid organ transplantation (18, 25). Common to both of these injuries is a localized response to injury in which injured endothelial and immune cells secrete growth and inflammatory cytokines, which elicits activation of normally quiescent medial vascular smooth muscle cells (VSMC) (18, 25). As part of this response to injury, VSMC migrate from the media into the lumen of the vessel, where they proliferate and synthesize cytokines to which they respond in an autocrine fashion, sustaining the progression of intimal hyperplasia. Some of the earliest receptor-initiated signaling events are mediated by the Rho family of small GTPases, which includes Rho, Rac, and Cdc42 (17, 31). The members of this family of signaling proteins act as molecular switches and cycle between inactive GDP-bound and active GTP-bound molecules to modify upstream signals to downstream effectors as required, and these are key regulators of cell proliferation, motility, and actin cytoskeleton assembly (17). The Rac proteins in particular play important roles in VSMC pathophysiology including regulation of oxidative processes, proliferation, and migration (1618, 28, 31).

Rac2 is a homologue of the ubiquitously expressed Rac1 GTPase that differs only in the COOH-terminal TRQQKRP amino acid motif (29). This motif is essential for Rac2-specific effects including membrane ruffling, myeloid differentiation, and cellular localization. Several important studies have demonstrated the significant functions of Rac2 in hematopoietic cells ranging from regulation of migration and cytoskeleton reorganization, maturation, oxidase activity, gene expression, adhesion, and host defense (10, 1416, 18, 22). Rac2 expression has long been considered to be restricted to hematopoietic cells, and in nonhematopoietic cells, these same functions have been ascribed to Rac1. Only one study describes the detection of Rac2 expression in VSMC (23). Aside from that report, no information regarding inducible Rac2 activation, expression, regulation of migration, and protein interactions has been described in VSMC.

Allograft inflammatory factor-1 (AIF-1) is a 143-amino acid, cytoplasmic, evolutionarily conserved, calcium binding protein. AIF-1 is constitutively expressed in inflammatory tissue and glial cells and has been implicated in the inflammatory process of several cell types, primarily macrophages and glial cells, and data from several groups in diverse systems advocate an important role for AIF-1 in inflammatory processes (8). These studies range from expression in infiltrating macrophages in rat cardiac allografts (30), microglial injury and activation (27), and in the allograft response of such phylogenetically distant species as carp and marine sponges (11, 21). AIF-1 has molecular signatures of a scaffold-signaling protein and interacts with actin but translocates to leading-edge lamellipodia in stimulated VSMC (3). We have previously shown that AIF-1 is not expressed in unstimulated VSMC but is rapidly expressed in response to injury and inflammatory cytokines (1). Overexpression of AIF-1 in VSMC results in increased proliferation and cell cycle protein expression, increased migration, and activation of Rac1 (2, 3).

A bacterial hybrid assay using AIF-1 as Bait determined Rac2 as a binding partner for AIF-1, which was confirmed by pull-down and coimmunoprecipitation in extracts from primary VSMC. We also investigated the functional significance of the AIF-1-Rac2 interaction. Rac2 expression and activity is believed to be restricted to hematopoietic cells, and considering the gap in our knowledge of Rac2 function in VSMC, we focused on characterization of Rac2 activity in VSMC, including its cytokine-inducible expression and activation. Furthermore, using retroviral overexpression, we have also determined the ramifications of Rac2 expression on VSMC migration, reactive oxygen species (ROS) generation, and signaling. The specific findings of this study indicate a functional interaction between AIF-1 and Rac2 in VSMC leading to Rac2 activation and a potential function for Rac2 in inflammation-driven VSMC activation. The synthesis of these data suggests the presence of an inflammation-responsive signal transduction pathway in VSMC mediated by AIF-1 interaction with Rac2.

MATERIALS AND METHODS

Cells and culture.

Primary rat aortic smooth muscle cells (RASMC) were obtained as cell explants from Sprague-Dawley rats and maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FBS as described (2). Cells were positive for SMC actin and negative for the primitive cell marker c-Kit, to rule out the remote possibility that a portion of cultured VSMC were derived from resident precursor cells that may be positive for Rac2, which is preferentially expressed in hematopoietic cells. Cells in passages 3–5 were used. Preconfluent VSMCs were serum-starved in 0.5% FCS for 48 h and then exposed to several, individual cytokines or growth factors: 15% FCS, 20% T cell-conditioned media, 100 U/ml interferon-γ (IFNγ), 20 ng/ml platelet-derived growth factor AB (PDGF-AB), 2 ng/ml transforming growth factor-β (TGFβ), 1 ng/ml tumor necrosis factor-α (TNFα), and 20 ng/ml interleukin-6 (IL-6). All cytokines were purchased from Sigma (St. Louis, MO); T cell-conditioned media was from Fisher Biotech.

Interaction studies.

Bacterial two-hybrid screen experiments were performed by using the Bacteriomatch two-hybrid system (Stratagene), using vectors, Escherichia coli strains, and methods supplied by the manufacturer. AIF-1 cDNA as Bait was cloned in-frame into pTRG or the modified pBT vector (pBTL) containing a Gly4-Ser3 linker. Cells containing the two-hybrid Bait plasmid pBT and prey plasmid pTRG were maintained on Luria broth (LB) agar containing chloramphenicol (34 μg/ml) or tetracycline (15 μg/ml). The LB selection agar for two-hybrid experiments contained 30 μg/ml carbenicillin, 15 μg/ml tetracycline, 34 μg/ml chloramphenicol, and 50 μg/ml kanamycin (i.e., CTCK), simultaneously selecting for protein-protein interaction and the presence of pBT and pTRG plasmids. Interactions were detected by growth and number of blue colonies and further verified by glutathione S-transferase (GST) pull-down assay. All constructs were verified by sequencing. The AIF-1/GST fusion protein pull-down and coimmunoprecipitation experiments were used to verify interactions and were performed as described (3). Anti-HA tag-Sepharose slurry was from Roche Diagnostics. Samples included a blocking peptide directed to the Rac2 protein COOH terminus (Santa Cruz).

Immunofluorescence.

Primary RASMC infected with HA-tagged AIF-1 adenovirus (AdAIF-1) were grown on chamber slides and incubated with T lymphocyte-conditioned media for 48 h to induce endogenous Rac2 expression. After serum starvation in 0.5% FCS for 24 h, some samples were stimulated with 20 ng/ml PDGF for 30 min. Cells were fixed in 10% formalin, blocked with 2% goat serum, and incubated with primary antibody for 1 h at room temperature, followed by a 30-min incubation with secondary antibody conjugated to AlexaFluor 568 (red) or AlexaFluor 488 (green) (Molecular Probes) and counterstained with 4,6-diamidino-2-phenylindole (DAPI) for nuclear staining. AIF-1 antibody was used at of 2 μg/ml, and Rac2 antibody was used at 1 μg/ml.

Retroviral construction and stable transduction.

The protein coding region of the mouse Rac2 cDNA was inserted into the retroviral packaging vector pLXSN containing the gentamicin resistance gene, and Rac2 retrovirus (Rac2-RV) was constructed using a kit from Clontech (La Jolla, CA) according to manufacturer’s instructions and as previously described (3). Supernatant containing recombinant high-titer virus was then used to infect rat VSMC in two 4-h exposures of viral supernatant in the presence of 8 μg/ml Polybrene (∼40–60% stable transduction was achieved). Stably transduced G418-resistant cells were pooled from each transduction to avoid the effects of clonal variation.

Rac activation was determined by the p21-activated kinase (PAK) pull-down assay, as described (3). The volume of lysate was normalized to protein concentration. Lysates were incubated with GST-PAK Sepharose beads (Cytoskeleton) for 1 h at 4°C. Only activated forms of Rac1 bind the PAK protein (17). Beads were washed three times, and bound proteins detected by Western blotting with Rac1 antibody and quantitated by densitometry.

Western blotting.

Cell extracts were prepared as described (4). Membranes were incubated with a 1:2,000 dilution of primary antibody and a 1:2,000 dilution of secondary antibody. Because of the very high sequence homology between Rac1 and Rac2 proteins, it was necessary to use anti-peptide antisera specific for the unique amino acids present on the COOH terminus of Rac2, as follows: Rac2 antibody (AbCam, Santa Cruz Biotechnology), monoclonal anti-Rac1 (Upstate Biotechnology), anti-phospho-Erk1/2-MAPK (Cell Signaling), anti-phospho-Pak1/2 (Biosource International), phospho-serine-144 and -141, of Pak1 and Pak2, respectively. Total Pak1, total p38, and phospho-p38 was from Cell Signaling, and monoclonal anti-GAPDH was from Biogenesis. AIF-1 antibody has been described (1). Equal protein concentrations of cell extracts were determined by Bradford assay, and equal loading on gels was verified by Ponceau S staining of the membrane. Reactive proteins were visualized using enhanced chemiluminescence (Amersham) according to the manufacturer’s instructions.

Migration and chemotaxis.

Transwell Boyden chamber plates (6.5 mm-diameter, Costar) with 8-μm polycarbonate membrane pore size were seeded with VSMC in medium containing 0.5% FCS as described (3). PDGF at 20 mg/ml was placed in the lower chamber, and cells were incubated for 3 h at 37°C, at which time cells were fixed and stained. VSMC that migrated to the lower surface of the membrane were quantitated by counting 4 high-powered fields per membrane. Experiments were performed in triplicate from three independently transduced groups of VSMC.

Intracellular generation of superoxide.

Conversion of nitroblue tetrazolium (NBT) to formazan was used as a measurement of intracellular superoxide generation (28). Equal numbers of RASMCs were serum-starved in 0.1% FCS for 48 h, then stimulated with 10% FCS for 1 to 24 h, followed by incubation with NBT (0.25 mg/ml in DMSO) for 1 h. After trypsinization, cells were counted, and pellets dissolved in DMSO and PBS. Absorbance was measured at 540 nm, and the NBT reduction to formazan was normalized to cell number.

Statistical analysis.

Data are means ± SE. The statistical significance regarding multigroup comparison was determined by two-way ANOVA with Bonferroni correction. A value of P < 0.05 was considered significant.

RESULTS

Interaction of Rac2 with AIF-1.

A protein-protein interaction screening based on a bacterial two-hybrid system identified an AIF-1-Rac2 interaction, which was verified in primary cultured VSMC by four different complementary techniques. First, extracts from primary rat VSMC isolated from uninjured arteries were incubated with an affinity resin consisting of a recombinant AIF-1-GST fusion protein immobilized to glutathione Sepharose beads or GST Sepharose beads as a negative control. The beads were washed, interacting proteins were separated by SDS-PAGE, and Rac2 was detected using a specific Rac2 antibody. Figure 1A shows that Rac2 protein pulls down with AIF-1. This interaction could be inhibited by addition of 200 ng of blocking peptide, which corresponds to a 10-amino acid sequence of COOH-terminal Rac2, a region which has been shown to be important for Rac2 protein interactions and cellular compartmentalization (29). Second, a nonfunctional AIF-1 mutant in which the EF-hand has been disrupted (4) was unable to interact with Rac2 in the GST pull-down assay, suggesting a functional significance for the AIF-1-Rac2 interaction. Third, Rac2 could be coimmunoprecipitated with AIF-1 in rat VSMC stably transduced with Rac2 retrovirus and infected with HA-tagged AdAIF-1 (Fig. 1C). AIF-1 interaction with Rac2 was further verified by colocalization in primary rat VSMC. Rat VSMC were infected with HA-tagged AdAIF-1, and endogenous Rac2 expression was induced by 48-h incubation of VSMC in T lymphocyte-conditioned media. After serum starvation in 0.5% FCS for 24 h, VSMC were then stimulated with PDGF. In unstimulated VSMC, AIF-1 and Rac2 colocalize in the cytoplasm (Fig. 1C). However, when VSMC are exposed to the chemotactic stimulus of PDGF for 30 min, a portion of the coimmunolocalization translocates to the lamellipodia. Together, these data corroborate the bacterial two-hybrid method screen demonstrating a Rac2-AIF-1 interaction, and translocation to the lamellipodia suggests that the AIF-1-Rac2 interaction has functional implications.

Fig. 1.

Interaction between Rac2 and allograft inflammatory factor (AIF-1) in vascular smooth muscle cells (VSMC). A: pull-down. Rat VSMC extracts were incubated with glutathione S-transferase (GST), AIF-1/GST, AIF-1ΔA/GST, or AIF-1/GST plus 200 ng Rac2 COOH-terminal peptide. Interacting Rac2 protein was identified by Western blot. Blot shown is representative of at least 3 independent experiments. B: coimmunoprecipitation. Rat VSMC that stably express empty vector or Rac2 were co-infected with HA-tagged AIF-1 adenovirus (AdAIF-1). AIF-1 was immunoprecipitated with anti-HA tag antibody, and precipitates were immunoblotted with anti-Rac2 or AIF-1 antibody. C: cellular colocalization. Rat VSMC infected with HA-tagged AdAIF-1 (green) were stimulated with T lymphocyte-conditioned medium to induce endogenous Rac2 expression (red). After serum starvation, VSMC were stimulated with platelet-derived growth factor (PDGF) for 30 min, at which time both Rac2 and AIF-1 translocate to lamellipodia. Magnification, ×400.

AIF-1 expression enhances Rac2 activation.

Based on the previous experiments, we hypothesized that AIF-1 expression modulated Rac2 activity. For these experiments, RASMC were infected with AdAIF-1 or green fluorescent protein adenovirus (AdGFP), serum-starved in 0.5% FCS for 48 h, then stimulated with PDGF. Cell extracts were incubated with GST-PAK beads, which bind only to activated forms of the Rac GTPase proteins (4). Figure 2 shows that while AdGFP-expressing VSMC demonstrate a typical Rac2 activation profile, AIF-1 expression can significantly increase Rac2 activation (P < 0.05). This is most notable in serum-starved, unstimulated cells, where Rac2 activation is enhanced an average of 193 ± 9.3% above AdGFP levels (n = 3).

Fig. 2.

Activation of Rac2 by AIF-1. Rat VSMC were incubated with T lymphocyte-conditioned media for 48 h to induce endogenous Rac2 expression. After serum starvation, cells were stimulated with 20 ng/ml PDGF, and activated, p21-activated kinase (PAK)-bound Rac2 was identified by Western analysis and quantitated by densitometry. One-tenth the amount of extract was also blotted for Rac2 and AIF-1. Gel shown is representative of 3 experiments performed on 3 independent groups of primary rat VSMC. Graph is a compilation of at least 3 different experiments using independent retrovirally transduced groups of VSMC. Errors are reported as standard error of mean (SE). *P < 0.05. AdGFP, green fluorescent protein adenovirus.

Rac2 is differentially expressed and activated in VSMC.

We have previously reported a relationship between AIF-1 expression and Rac1 activation (3). Since Rac2 expression and function in vascular cells has not been fully described, we focused our efforts on the characterization of Rac2 in VSMC. Rac2 expression and activity is believed to be restricted to hematopoietic cells, and the next series of experiments were designed to characterize its expression and activation in VSMC and compare this expression with Rac1. For these experiments, cultured rat VSMC were serum-starved in 0.5% FCS for 48 h, and then stimulated with a serum or a variety of cytokines for 48 h to induce differential protein expression. The results presented in Fig. 3 demonstrate a detectible level of Rac2 protein expression in unstimulated VSMC. IFNγ, PDGF, and IL-6, demonstrated no significant increase in Rac2 expression, whereas T cell-conditioned media, FCS, TNFα, and TGFβ elicited a 371 ± 81%, 213 ± 41%, 248 ± 38%, and 255 ± 61% increase over unstimulated cells, respectively, and were significantly different than unstimulated VSMC in all experiments (P < 0.05 for 3 experiments). In contrast, the much more robust Rac1 levels were essentially unchanged between unstimulated and cytokine stimulated VSMC.

Fig. 3.

Differential expression of Rac2 protein by cytokines in primary rat aortic smooth muscle cells (RASMC). Cells were serum-starved for 48 h, then stimulated with the indicated cytokines. Equal concentrations of protein extracts were probed with Rac2 antibody. For comparison, blots were then stripped and blotted with anti-Rac1 antibody. Rac protein expression was quantitated by densitometry. Blot and densitometry shown are representative of at least 3 similar experiments from 3 independently stimulated groups of VSMC. IFNγ, interferon-γ; TGFβ, transforming growth factor-β; TNFα, tumor necrosis factor-α; IL-6, interleukin-6; C.M., conditioned medium.

Cytokine-mediated Rac2 activation was assessed by the Pak pull-down assay. Cultured rat VSMC were serum-starved in 0.5% FCS for 48 h and then stimulated with a serum or a variety of cytokines. Figure 4 shows that of every factor tested, only 15% FCS and T cell-conditioned media, were capable of significantly inducing Rac2 activity over unstimulated cells (P < 0.05). This is similar to that of Rac1, in which 15% FCS, T cell-conditioned media, and also PDGF were the most effective inducers of Rac1 activity. Together, these experiments are the first to demonstrate differential expression and activation of Rac2 in VSMC by serum and cytokines.

Fig. 4.

Differential activation of Rac2 by cytokines in primary RASMC. VSMC were incubated with T lymphocyte-conditioned media for 48 h to induce endogenous Rac2 expression, serum-starved for 48 h, then stimulated with the indicated cytokines for 5 min. Equal concentrations of protein were incubated with GST-PAK Sepharose beads, and activated, PAK-bound Rac2 was identified by Western analysis. For comparison, blots were stripped and blotted with anti-Rac1 antibody. Rac2 and Rac1 activation was quantitated by densitometry and is representative of at least 3 similar experiments from 3 independently stimulated groups of VSMC. Errors are reported as standard error of mean (SE). *P < 0.05 for Rac2 stimulated vs. unstimulated.

Expression of Rac2 increases VSMC migration.

The function of Rac2 in VSMC is uncharacterized. Based on its demonstrated role in hematopoietic cell migration, we investigated the effects of Rac2 expression on VSMC migration. To attribute the effects exclusively to Rac2, without a confounding background of external activating stimuli, it was necessary to overexpress Rac2. Primary rat VSMC were stably transduced with Rac2 retrovirus containing the protein coding region of Rac2 cDNA (Rac2-RV) or with vector alone (empty RV) and stable transductants selected with antibiotics. Individual colonies were combined to avoid any effects of clonal selection, and constitutive Rac2 protein expression was verified by Western blot (Fig. 5A). Retrovirally expressed Rac2 levels were similar to endogenous levels induced by conditioned media. To demonstrate an effect on migration, stably transduced VSMC were seeded into Boyden chambers, and differences in chemotaxis were quantitated by counting cells that migrated in response to PDGF. Figure 5B shows that VSMC stably transduced with Rac2 migrate 128% more rapidly (129.86 ± 13.84 vs. 57.03 ± 4.50 for Rac2 and controls, respectively; P < 0.05) than control cells, even in the absence of chemotactic stimuli. In the presence of 20 ng PDGF, Rac2 cells migrate 72% more rapidly (352.31 ± 41.10 vs. 204.50 ± 24.00 for Rac2 and control, respectively; P < 0.001), indicating a positive effect of Rac2 expression on VSMC migration. These data are the first to demonstrate a functional consequence of Rac2 expression in VSMC.

Fig. 5.

Rac2 expression enhances VSMC migration. A: retroviral mediated overexpression of Rac2. Extracts from unstimulated, empty vector retrovirus (“Empty RV”), Rac2 RV, and conditioned media treated VSMC were Western blotted with Rac2 antibody. B: VSMC stably transduced with Rac2 or empty vector were seeded onto Boyden chamber membranes and exposed to 20 ng/ml PDGF for 3 h. Some samples remained unstimulated. Migrating VSMC were quantitated by counting 4 high-powered fields (HPF) per membrane. C: activation of PAK by Rac2. Extracts from stably transduced, serum-starved VSMC were Western blotted with anti-phospho-Pak1/2 antibody, or total Pak1, or GAPDH antibody as protein loading controls. Values are means from 3 experiments performed in triplicate from 3 independent retrovirally transduced groups of VSMC. Errors are reported as standard error of mean (SE). *P < 0.05. **P < 0.001.

Migration is regulated by the activation of Rac1 and its effectors. The Pak proteins are auto-phosphorylated on serine-144 and serine-141, (for Pak1 and Pak2, respectively) in its activated form, by association with the Rac proteins (3, 4, 19, 32). The ability of Rac2 to induce migration in the absence of stimuli led us to speculate that Rac2 was constitutively activated, which would be detected by activated, phosphorylated Pak1. Extracts from serum-starved Rac2 and empty vector-expressing VSMC were immunoblotted with phospho-specific Pak1/2 antibody, which recognizes phosphoserine-144 or -141, which are located at identical residues, for Pak1 and Pak2, respectively. Very little Pak2 was detectible in our rat VSMC, so total Pak1 was used for normalization. Figure 5C shows that Pak1 is constitutively phosphorylated in Rac2-expressing VSMC, indicating that overexpressed Rac2 is active even in the absence of serum growth factors.

Expression of Rac2 increases intracellular superoxide production.

In hematopoietic cells, Rac2 is known to be required for assembly and activation of the NADPH oxidase complex and generation of ROS. We hypothesized a similar function for Rac2 in VSMC. To assess a role for Rac2 in generation of ROS in VSMC, superoxide production was examined in serum-stimulated VSMC. Superoxide generation was measured by NBT reduction (28). Rac2-expressing cells produced significant increases in 10% FCS-induced superoxide production at 1 h, 2 h, and 24 h after FCS stimulation, suggesting that Rac2 expression can enhance superoxide generation in VSMC (Fig. 6).

Fig. 6.

Rac2 expression increases intracellular superoxide production in VSMC. Rat VSMC stably transduced with Rac2 or empty vector were serum-starved in 0.1% FCS for 48 h, followed by stimulation with 10% FCS. Nitroblue tetrazolium was added to cells for 1 h, and intracellular accumulation of formazan was measured at 540 nm. Control empty vector VSMCs showed a time-dependent superoxide production. +P < 0.05 compared with unstimulated cells with empty vector. Rac2-expressing cells produced a significant increase in superoxide production at all time points tested stimulation. *P < 0.05 and **P < 0.01 compared with empty vector VSMC.

Expression of Rac2 increases activation of Erk1/2 and p38 MAP kinases.

In hematopoietic cells, activation of Rac2 has demonstrated effects on modulation of downstream signaling cascades, and we hypothesized this would be similar in VSMC. Primary rat VSMC stably transduced with empty vector or Rac2-RV were serum-starved in 0.5% FCS for 48 h, then stimulated with PDGF for different times. Activations of cytoplasmic kinases in extracts from these cells were quantitated by Western blot using phospho-specific antibodies. Figure 7 shows that activation of Erk1/2 MAPK and p38 MAPK are increased by the expression of Rac2. Phosphorylation of both Erk1/2 and p38 is significantly increased at 5 min after PDGF stimulation (P < 0.05).

Fig. 7.

Rac2 modulates signaling pathways in VSMC. Rat VSMC stably transduced with Rac2 or empty vector were serum-starved for 48 h, then stimulated with 20 ng/ml PDGF for the times indicated. A: equal amounts of protein extracts were immunoblotted for phospho-ERK1/2 or phospho-p38. The blot was then stripped and incubated with anti-total ERK1/2, p38, and GAPDH antibody to indicate equal protein loading. There was no significant difference between unstimulated and 5-min PDGF-stimulated Rac2-expressing cells. B: ERK1/2 activation was quantified by densitometry. *P < 0.05, for Rac2 compared with empty vector. C: p38 activation was quantified by densitometry. *P < 0.05, for Rac2 compared with empty vector at 5 min. †P < 0.057 for unstimulated VSMC. Blot shown is representative of 3 independently stimulated groups, with densitometry from 3 groups.

DISCUSSION

A two-hybrid screen using AIF-1 as Bait determined Rac2 as a binding partner for AIF-1, which was confirmed by pull-down assay in extracts from primary VSMC. This interaction was unexpected in that unlike Rac1, which is ubiquitously expressed, Rac2 expression had been considered to be restricted to cells of hematopoietic origin (14, 22). Since the VSMC were isolated from normal, uninjured arteries, the likelihood that these VSMC are a progenitor population more similar to hematopoietic cells than to vascular smooth muscle is very remote. A nonfunctional mutant of AIF-1 was impaired in its ability to bind to Rac2, and the AIF-1-Rac2 interaction was abrogated by addition of the TRQQKRP COOH-terminal Rac2 peptide, a unique region essential to Rac2-specific function (29). The 93% amino acid homology between Rac1 and Rac2 also necessitates use of antisera specific for this unique COOH-terminal region, rather than antisera raised against the entire protein. Furthermore, the very high degree of identity at the nucleic acid level among Rac family proteins limits the efficacy of siRNA knockdown approaches to specifically inhibit Rac2 expression. We have previously shown that AIF-1 resides in the cytoplasm and translocates to lamellipodia upon PDGF stimulation (3). Similarly, Rac2 and AIF-1 colocalized in the cytoplasm of rat VSMC and cotranslocated to lamellipodia upon PDGF stimulation. Expression of AIF-1 in rat VSMC also increased Rac2 activation. Together, these data indicated a functional interaction of the AIF-1-Rac2 binding, most likely involving cellular locomotion. We have previously reported that AIF-1 polymerizes actin and increases Rac1 activation (3). Patterson el al. (23) reported Rac2 expression in thrombin-treated VSMC. However, since Rac2 expression and function in vascular cells has not been explored, we considered it of greater impact to characterize Rac2 expression and activity in VSMC, rather than define the mechanistic relationship between AIF-1 and Rac2.

Because of the gap in our knowledge concerning Rac2 function in VSMC, we designed experiments to investigate differential expression and activation of Rac2 by cytokines in VSMC. In cytokine-stimulated VSMC, inflammatory factors such as FCS and T lymphocyte-conditioned media, which contain multiple inflammatory mediators, elicited the most robust expression of Rac2 compared with other factors. This expression pattern of Rac2 is similar to AIF-1, which is also constitutively expressed in leukocytes and only expressed in VSMC when they are cytokine-stimulated (1). In contrast, the expression of Rac1 is more ubiquitous and remains unchanged regardless of stimulation. In a similar manner, Rac2 activation was also cytokine-responsive. FCS and T cell-conditioned media, both of which contain multiple inflammatory and growth factors, elicited the strongest activation of Rac2. TGFβ and TNFα also significantly increased Rac2 expression. Other individual cytokines tested did not significantly activate Rac2. Conversely, Rac1 activation in VSMC was responsive to all of the cytokines and growth factors tested. Together these data suggest that in VSMC, Rac2 expression and activation are driven by inflammatory stimuli and implicate a role for Rac2 in vascular diseases including restenosis and atherosclerosis, which are mediated by inflammation. This also suggests that Rac2 activity is not a redundant homologue of Rac1 but has functional specificity distinct from Rac1.

Similar to Rac1, Rac2 is involved in chemotaxis, actin remodeling, and generation of reactive oxygen in neutrophils, but there are some specific differences in gene expression, cellular localization, signaling, and gene expression between Rac1 and Rac2 (10, 1416, 18). In our experiments, overexpression of Rac2 significantly increased both basal and PDGF-driven VSMC migration, and this is the first report of Rac2 function in VSMC. There were significantly increased levels of phosphorylated serine-144 Pak1 in Rac2-expressing cells. Pak1 autophosphorylation at this residue is directly responsive to Rac or Cdc42 interactions (19, 32) and may account for the increased VSMC migration seen in Rac2 VSMC. Rac2 effects on VSMC migration suggests an important role for Rac2 in VSMC migration in response to inflammation.

Components of NADPH oxidase are also present in VSMC, and NADPH oxidase-derived ROS is important for the regulation of vessel tone as well as the pathophysiology of many vascular diseases (27). Deficits in superoxide production, actin polymerization, and p42/44 activation were noted in Rac2 neutrophils from Rac2 deficient mice (25). Patterson et al. (23) also demonstrated the increased expression of Rac2 protein in VSMC by thrombin treatment, which correlates with ROS generation. Overexpression of Rac2 in VSMC indicated a significant high intracellular superoxide production by FCS treatment characterized by a relatively slow and prolonged ROS production, which was quite different from the faster and greater response in observed in phagocytes (12). Association of Rac2 expression and superoxide generation provides additional support for the involvement of VSMC-derived Rac2 in ROS-mediated vascular diseases.

Rac2 has been implicated in the regulation of the cross-cascade activation between phosphoinositide 3-kinase and classical p21ras-Raf-Mek-Erk pathway in mast cells (13). Rac2−/− neutrophils showed marked defects in cell migration and severely reduced Erk phosphorylation (14). In our studies, overexpression of Rac2 resulted in a significant enhancement of Erk1/2 MAPK activation at 5 min after PDGF stimulation. While it has been shown that Rac1 binds to and stimulates the kinase activity of Pak1 more efficiently than Rac2 does (20), we were able to show that in VSMC, constitutive Rac2 expression can also significantly activate Pak, even in serum-starved conditions. This activation in the absence of stimuli is similar to what we observed for Erk1/2 phosphorylation and VSMC migration in the absence of chemotactic stimuli. It may be likely that the constitutive Rac2 activation is the mechanism whereby migration and Erk1/2 activation can occur in the absence of stimuli. Not only are the Pak kinases downstream effectors for the small GTPases, but these are also proximal to p38 activation (5, 7, 33). The activation of p38 in Rac2-expressing VSMC is significant in that the production and activation of proinflammatory proteins is often mediated by p38 MAP kinase, and sustained activation of p38 is an important contributor to the vascular response to injury.

This report details a functional interaction between AIF-1 and Rac2 in VSMC. It also shows, for the first time, that Rac2 is differentially expressed and activated in VSMC in response to inflammatory cytokines and that its expression and activation differs from the ubiquitously expressed Rac1. Overexpression of Rac2 increases intracellular superoxide production, increases VSMC migration, and enhances activation of its downstream effector Pak1, as well as other distal effectors, Erk1/2 and p38. Together, these data suggest an important function for Rac2 in inflammation-driven VSMC activation and afford insight into the mechanisms by which Rac2 participates in the pathogenesis of vascular diseases.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant HL-63810 and American Heart Association Grant 0455562U to M. V. Autieri. Y. Tian is the recipient of American Heart Association Predoctoral Fellowship 0515320U.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract