Abstract
Monocyte/macrophage infiltration to the subendothelial space of arterial wall is a critical initial step in atherogenesis, in which CC chemokine ligand 2 (CCL2)/monocyte chemoattractant protein-1 (MCP-1) is thought to play a key role. This study investigated the effectiveness of phosphodiesterase inhibitors, including the nonselective pentoxifylline (PTX) and the selective type III (cilostamide) and type IV (denbufylline) inhibitors, on cytokine-induced CCL2/MCP-1 production in cultured rat vascular smooth muscle cells (VSMCs), and the signal transduction mechanisms whereby they act. Our results showed that tumor necrosis factor (TNF)-α induced a marked increase in CCL2/MCP-1 production in dose- and time-dependent manners. 2-(2-Amino-3-methoxyphenyl)-4H-1-benzopyran-4-one (PD98059), 1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio) butadiene (U0126) [both inhibitors of p42/44 mitogen-activated protein kinase (MAPK) kinase], and anthra[1hyphen]9-cd]pyrazol-6(2H)-one (SP600125) [an inhibitor of c-Jun NH2-terminal kinases (JNKs)] attenuated TNF-α-induced CCL2/MCP-1 production, without affecting I-κBα degradation or p65/nuclear factor-κB (NF-κB) nuclear translocation. PD98059 abolished TNF-α-activated p42/44 MAPK phosphorylation and c-Fos up-regulation, whereas SP600125 inhibited TNF-α-activated JNK and c-Jun phosphorylation. The NF-κB inhibitor carbobenzoxy-l-leucyl-l-leucyl-l-leucinal (MG132) attenuated TNF-α-induced CCL2/MCP-1 production in the presence of increased phospho-JNK and phospho-c-Jun levels. When SP600125 was added simultaneously, MG132 completely inhibited TNF-α-induced CCL2/MCP-1 production. Finally, the pretreatment of VSMCs with PTX or cilostamide, but not denbufylline, reduced TNF-α-induced CCL2/MCP-1 production, which was preceded by attenuation of p65/NF-κB nuclear translocation, p42/44 MAPK, and JNK-c-Jun phosphorylation, and c-Fos up-regulation. These data indicate that TNF-α-stimulated CCL2/MCP-1 production in rat VSMCs is dually regulated by activator protein-1 (AP-1) and NF-κB pathways, and inhibition of type III phosphodiesterase contributes substantially to the suppressive effect of PTX on CCL2/MCP-1 production via down-regulation of AP-1 and NF-κB signals.
Monocyte infiltration and accumulation in the subendothelial space of the arterial wall is a prominent pathobiological feature in early atherogenesis (Ross, 1999), in which chemokines are thought to play a key role (Reape and Groot, 1999; Gerszten et al., 2000). Among these, the cytokine-induced CCL2/monocyte chemoattractant protein-1 (MCP-1) is noteworthy for its ability to promote migration of monocytes harboring its cognate receptor, chemokine receptor 2 (CCR2) (Reape and Groot, 1999). A growing body of evidence indicates that local overexpression of CCL2/MCP-1 by infiltrating monocytes or vascular cells induces accumulation of monocytes/macrophages and formation of atherosclerotic lesion, which seems to synergize with hypercholesterolemia (Ikeda et al., 2002; Namiki et al., 2002). Parallel to these findings, expression of CCR2 has also been documented in experimental atherosclerotic lesions in cholesterol-fed rabbits (Ohtsuki et al., 2001). Further in vivo studies have shown that genetically altered mice deficient for CCR2 and CCL2/MCP-1 are protected from development of atherosclerosis (Peters and Charo, 2001) and that anti-CCL2/MCP-1 gene therapy attenuates atherosclerosis in apolipoprotein E-knockout mice (Ni et al., 2001). In humans, CCL2/MCP-1 polymorphism (2518 G/G high MCP-1 producer genotype) has been implicated with elevated lipoprotein(a) levels and an increased susceptibility to coronary artery disease (Szalai et al., 2001). Together, these data support a critical role of the MCP-1/CCR2 system for atherogenesis.
CCL2/MCP-1 was originally discovered from mononuclear leukocytes (Robinson et al., 1989; Yoshimura et al., 1991) but is now known to be produced by many nonimmune cells, including vascular smooth muscle cells (VSMCs) (Ortego et al., 1999; de Keulenaer et al., 2000; Iseki et al., 2000). Its expression is regulated by a number of stimuli, most notably the proinflammatory cytokine tumor necrosis factor (TNF)-α (Libby and Galis, 1995). Many of the cellular responses induced by TNF-α require cytoplasmic signals transduced to two major transcription factors, nuclear factor-κB (NF-κB) and activator protein-1 (AP-1) (Karin, 1995; Kyriakis, 1999; Leong and Karsan, 2000; Baud and Karin, 2001). NF-κB consists of a homoor heterodimer formed from five possible subunits of the Rel family, with the p65/p50 heterodimer being the most common form. Its activation by TNF-α is associated with phosphorylation of inhibitory protein of NF-κB (I-κB) by I-κB kinase complexes, degradation of the phosphorylated I-κB, and translocation of the freed NF-κBto the nucleus, which triggers proinflammatory and antiapoptotic gene expression (Leong and Karsan, 2000; Baud and Karin, 2001). By contrast, activation of AP-1 upon TNF-α treatment is regulated by three of the five known mammalian mitogen-activated protein kinases (MAPKs), the p42/44 MAPK, p38 MAPK, and c-Jun NH2-terminal kinase (JNK). Upon stimulation, these protein kinases enter the nucleus to induce or phosphorylate subunits of AP-1, including Jun and Fos proteins. The resultant enhanced AP-1 activity can then participate in the regulation of genes involved in inflammation and cell survival (Karin, 1995; Kyriakis, 1999). In cultured rat VSMCs, TNF-α has been shown to induce CCL2/MCP-1 expression via activation of the NF-κB signals (Ortego et al., 1999; Iseki et al., 2000), and the p42/44 MAPK pathways (de Keulenaer et al., 2000). However, information is limited regarding the role of AP-1 and other MAPK members in TNF-α-dependent CCL2/MCP-1 expression in VSMCs. We hypothesize that TNF-α may act on multiple kinase cascades that converge to AP-1 or NF-κB signals and activate CCL2/MCP-1 expression. Thus, measures capable of modulating these signaling pathways and the resultant CCL2/MCP-1 production may have therapeutic value in the prevention and treatment of atherosclerotic vascular diseases.
Recently, we have shown that pentoxifylline (PTX), a nonselective phosphodiesterase (PDE) inhibitor, is capable of suppressing the production of CX3CL1/fractalkine in TNF-α-activated VSMCs via down-regulation of the p42/44 MAPK and NF-κB signals (Chen et al., 2003). In VSMCs, there exist at least four PDE isozymes (types I, III, IV, and V) (Rabe et al., 1994; Pauvert et al., 2002). Among these, the types III and IV PDE isozymes hydrolyze only cAMP, the type V PDE hydrolyzes only cGMP, and the type I PDE accepts both nucleotides as a substrate (Dousa, 1999). Based on our previous work that the incubation of VSMCs with PTX leads to an increase in intracellular cAMP but not cGMP levels (Chen et al., 1999), it is speculated that PTX acts predominantly by inhibiting type III or IV PDE in VSMCs. In this study, we investigated 1) the role of the MAPK, NF-κB, and AP-1 pathways in TNF-α-dependent CCL2/MCP-1 production; 2) the effectiveness of PTX, cilostamide (a selective type III PDE inhibitor), and denbufylline (a selective type IV PDE inhibitor) on TNF-α-induced CCL2/MCP-1 production; and 3) the underlying signal transduction pathways whereby these PDE inhibitors act.
Materials and Methods
Reagents. Dulbecco's modified Eagle's media (DMEM), penicillin/streptomycin, fetal calf serum (FCS), and other tissue culture reagents were purchased from Invitrogen (Carlsbad, CA). The FCS was heat-inactivated before use. Culture flasks and plates were purchased from Costar (Cambridge, MA). Cilostamide and PTX were purchased from Sigma-Aldrich (St. Louis, MO). Denbufylline, PD98059, U0126, SB203580, and MG132 were obtained from Calbiochem (La Jolla, CA). SP600125 was purchased from Tocris Cookson Inc. (Avonmouth, UK). Recombinant rat TNF-α was obtained from R&D Systems (Minneapolis, MN). Rabbit anti-rat MCP-1 was purchased from PetroTech EC Ltd. (London, UK). Rabbit anti-p44/42 MAPK and mouse anti-phosphorylated p44/42 MAPK were obtained from New England BioLabs (Beverly, MA). Mouse anti-phosphorylated JNK and phosphorylated c-Jun, and rabbit anti-JNK and anti-c-Jun, anti-c-Fos, anti-p65/NF-κB, and anti-I-κBα were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Mouse anti-β-actin was obtained from Sigma-Aldrich. All chemicals used for total RNA isolation, reverse transcription-polymerase chain reaction, Northern blot analysis, whole cell lysate extraction, and Western blot analysis were of molecular grade and were obtained from Sigma-Aldrich or Roche Diagnostics (Mannheim, Germany) unless otherwise specified.
Cell Culture and Experimental Conditions. Primary culturing of rat aortic VSMCs was performed and characterized as described previously (Chen et al., 1999). Briefly, the thoracic aortas were removed from male Sprague-Dawley rats immediately after the rats were decapitated. The endothelium and adventitia were removed gently by scratching with forceps. Specimens were cut into 5-mm rings and digested in Hanks' balanced salt solution containing 15 mM HEPES, 0.2 mM CaCl2, 1 mg/ml collagenase, 0.125 mg/ml elastase, 0.375 mg/ml soybean trypsin inhibitor, and 2 mg/ml bovine serum albumin at 37°C for 90 min. The digested tissue was strained through a 180-μm steel mesh. After centrifugation at 200g for 5 min, the cells were resuspended in DMEM containing 20% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin. The cells were then seeded on flasks at a density of 1 × 104 cells/cm2 and incubated at 37°C in a humidified 5% CO2, 95% O2 air atmosphere. Cells were confirmed to be VSMCs by morphological criteria, by the presence of smooth muscle α-actin staining, and by the absence of staining for cytokeratin and factor VIII-related antigen with the avidin-biotin-peroxidase method. Cells between 10 and 20 passages were used and grown in DMEM containing 10% FCS. To determine the regulatory role of various MAPKs on TNF-α-stimulated CCL2/MCP-1 expression by VSMCs, cells were first grown in DMEM containing 10% FCS until reaching 90% confluence. The medium was then replaced by DMEM containing 0.5% FCS for 24 h before treatment with a p44/42 MAPK kinase inhibitor PD98059 or U0126 (10–40 μM and 5–20 μM, respectively, for 30 min), a p38 MAPK inhibitor SB203580 (5–20 μM for 30 min), or a broad-spectrum JNK inhibitor SP600125 (5–20 μM for 30 min). After preincubation, cells were stimulated with TNF-α (5 ng/ml) for 4 or 24 h. Further experiments were conducted to examine the role of NF-κB and AP-1 on TNF-α-stimulated CCL2/MCP-1 expression by VSMCs. The NF-κB inhibitor MG132 (5–20 μM) and the c-Jun/AP-1 inhibitor SP600125 (5–20 μM) were incubated with VSMCs for 1.5 h and 30 min, respectively. Then, cells were stimulated with TNF-α (5 ng/ml) for 7.5 to 120 min for Western blot analysis (signal transduction pathways) and immunocytochemistry (nuclear p65/NF-kB), for 4 h for Northern blot analysis (CCL2/ MCP-1 mRNA), or for 24 h for Western blot analysis (CCL2/MCP-1 protein).
Additional studies were designed to examine the role of PDE inhibitors in VSMC CCL2/MCP-1 expression. Cells were preincubated with PTX (0.5–2 mM) for 45 min, or cilostamide (10–40 μM) or denbufylline (10–40 μM) for 30 min, followed by TNF-α (5 ng/ml) for 7.5 to 120 min, 4 h, or 24 h. After incubation for the given periods, cell monolayers were washed and used for Northern blot analysis (4-h stimulation) or Western blot analysis (7.5- to 120-min or 24-h stimulation).
In all experiments mentioned above, corresponding controls were performed by incubating VSMCs with vehicles [either dimethyl sulfoxide or phosphate-buffered saline (PBS)] wherever appropriate.
Northern Blot Analysis. Total RNA was extracted by the acid guanidinium thiocyanate phenol chloroform method as described previously (Chen et al., 1999). Ten micrograms of total RNA were electrophoresed on formaldehyde-denatured 1% agarose gels and subsequently transferred to nylon membranes. To synthesize riboprobes for Northern blot hybridization, cDNA fragments of rat CCL2/MCP-1, c-jun, c-fos, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were first amplified by reverse transcription-polymerase chain reaction from glomerular RNA of Wistar rats using the following specific primer pairs: rat CCL2/MCP-1, upstream 5′-TCAGCCAGATGCAGTTAATG-3′ and downstream 5′-TTCTCTGTCATACTGGTCAC-3′ (GenBank accession no. M57441); rat c-jun, upstream 5′-TTCTGAAGCAGAGCATGACC-3′ and downstream, 5′-TTGAAGTTGCTGAGGTTGGC-3′ (GenBank accession no. X17163); rat c-fos, upstream, 5′-GCCTTTCCTACTACCATTCC-3′ and downstream, AGTTGATCTGTCTCCGCTTG-3′ (GenBank accession no. X06769); and rat GAPDH, upstream 5′-tcattgacctcaactacatg-3′ and downstream 5′-caaagttgtcatggatgacc-3′ (GenBank accession no. NM_017008). Reverse transcription-polymerase chain reaction was performed as described previously (Chen et al., 2003). The amplified products were eluted from polyacrylamide gels and subcloned into pGEM-T vectors (Promega, Madison, WI). The accuracy of the inserts were confirmed by sequence analysis, and the cloned cDNAs were linearized and used as templates for in vitro transcription of antisense digoxigenin-conjugated riboprobes, following the manufacturer's instructions (Roche Diagnostics). After hybridization, the blots were developed using CSPD (Roche Diagnostics) as the substrate for alkaline phosphatase. The intensity of the signal was then quantified with computerized densitometry, and normalized against the signal of GAPDH messages.
Western Blot Analysis. VSMCs were washed with ice-cold 1× PBS and lysed at 4°C for 15 min in lysis buffer (pH 7.4) containing 50 mM Tris HCl, 150 mM NaCl, 1% Igepal CA-630 (Sigma-Aldrich), 0.25% sodium deoxycholate (Sigma-Aldrich), 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, and 1 μg/ml each aprotinin, leupeptin, and pepstatin. Thirty micrograms of cell lysates was heated at 100°C for 10 min, applied to 9% SDS-polyacrylamide gels, and electrophoresed for detection of p44/42 MAPK, JNK, c-Jun, c-Fos, IκBα, and β-actin. To detect CCL2/MCP-1, the conditioned media of TNF-α-activated VSMCs were concentrated with Centricon-10 (Millipore, Bedford, MA), and 20 μg of protein was electrophoresed on 12% SDS-polyacrylamide gels. A prestained marker was also electrophoresed as a molecular weight marker. After electrophoresis, proteins were transferred onto a polyvinylidene difluoride membrane (Millipore) using a transblot chamber with Tris buffer. Western blots were incubated at 4°C overnight with primary antibodies. The next morning, membranes were washed with 1× PBS/5% Tween 20 at room temperature for 40 min, and incubated with peroxidase-conjugated second antibodies at room temperature for 1 h. After washing, the membranes were incubated with Renaissance (PerkinElmer Life Sciences, Boston, MA) according to the manufacturer's instructions. The intensity of the signal was then quantified with computerized densitometry and normalized against the signal of β-actin.
Immunocytochemistry. To demonstrate p65/NF-κB nuclear translocation, VSMCs, with or without pretreatment with the given inhibitors, were incubated with TNF-α (5 ng/ml) or vehicles for 7.5, 15, or 30 min before fixation with 4% paraformaldehyde for 1 h at 4°C. The cells were then washed by 1× PBS/0.2% Triton X-100 for 15 min and incubated with a rabbit anti-p65/NF-κB antibody at 4°C overnight. The next day, after washing for 15 min, the cells were incubated with biotin-conjugated anti-rabbit IgG at room temperature for 1 h. Then, the cells were washed and incubated with the avidin-biotin-peroxidase reagent (Dakopatts, Glostrup, Denmark) at room temperature for 1 h. After washing, immunodetection for p65/NF-κB was performed by adding 3-amino-9-ethylcarbazole chromogen as substrate according to the manufacturer's instructions.
Statistics. Data are expressed as mean ± S.E.M. All comparisons were done by analysis of variance using the StatView package for the Macintosh computer (Abacus Concepts, Berkeley, CA). A probability value of less than 0.05 was considered statistically significant.
Results
Induction of CCL2/MCP-1 mRNA and Protein Expression by TNF-α. The incubation of VSMCs with different concentrations of TNF-α (0.5–50 ng/ml) for varying time periods (2–24 h) resulted in marked up-regulation of a single 0.9-kilobase CCL2/MCP-1 mRNA species and approximately 16- to 30-kDa CCL2/MCP-1 proteins in time- and dose-dependent manners (Fig. 1). The presence of several molecular masses of CCL2/MCP-1 protein probably represent different degrees of glycosylation as reported previously (Satriano et al., 1993; Rovin et al., 1994).
Roles of MAPKs in TNF-α-Induced CCL2/MCP-1 mRNA and Protein Expression. To examine the roles of p42/44 MAPK, p38 MAPK, and JNK in TNF-α-dependent CCL2/MCP-1 production, VSMCs were incubated with PD98059, an inhibitor of p42/44 MAPK kinase (Means et al., 2000); SB203580, an inhibitor of p38 MAPK (Davies et al., 2000); or SP600125, an inhibitor of JNK (Bae and Song, 2003). The results showed that PD98059 (10–40 μM) and SP600125 (5–20 μM) dose dependently attenuated TNF-α-stimulated CCL2/MCP-1 mRNA induction (Fig. 2A). The incubation of VSMCs with another potent inhibitor of p42/44 MAPK kinase, U0126, also resulted in dose-dependent (5–20 μM) attenuation of MCP-1 gene and protein expression (data not shown). When VSMCs were pretreated with both PD98059 and SP600125 in combination, TNF-α-induced CCL2/MCP-1 mRNA expression was additively inhibited (Fig. 2B). These effects of PD98059 and SP600125 were also confirmed at the protein level (Fig. 2B). By contrast, SB203580 (5–20 μM) did not have discernible effects on TNF-α-induced CCL2/MCP-1 mRNA expression (Fig. 2A).
Additional experiments were performed to examine the differential effects of PD98059 and SP600125 on TNF-α-activated signaling cascades. The pretreatment of PD98059 (20 μM) abolished TNF-α-activated p42/44 MAPK phosphorylation and c-Fos up-regulation without affecting JNK-c-Jun phosphorylation, I-κBα degradation, or p65/NFκB nuclear translocation (Fig. 3, A and C). The preincubation of VSMCs with U0126 (10 μM) resulted in a similar pattern of inhibition in TNF-α-activated signaling pathways as did PD98059 (data not shown). By contrast, the pretreatment of VSMCs with SP600125 (10 μM) completely inhibited TNF-α-activated phosphorylation of JNK2/3 and c-Jun, without affecting p42/44 MAPK phosphorylation, c-Fos up-regulation, I-κBα degradation, or p65/NFκB nuclear translocation (Fig. 3, B and C).
Roles of NF-κB and c-Jun/AP-1 in TNF-α-Induced CCL2/MCP-1 mRNA and Protein Expression. The ubiquitin/proteosome inhibitor MG132 (Nakayama et al., 2001) was used to examine the role of NF-κB in TNF-α-dependent CCL2/MCP-1 production. Figure 4A shows that preincubation of VSMCs with MG132 (5–20 μM) dose dependently attenuated TNF-α-stimulated CCL2/MCP-1 mRNA expression (Fig. 4A). When VSMCs were preincubated with SP600125 (10 μM) and MG132 (20 μM) in combination, TNF-α-induced CCL2/MCP-1 production was completely inhibited (Fig. 4, B and C). At the concentrations used in this study, our previous work has shown that MG132 abolishes TNF-α-induced p65/NF-κB nuclear translocation in the presence of an augmented phospho-c-Jun level (Chen et al., 2003). This study further demonstrates that MG132 augmented TNF-α-induced JNK2/3 phosphorylation without affecting c-Fos up-regulation (Fig. 5). These results are consistent with our current knowledge that MG132, besides antagonizing NF-κB activity, also activates the JNK-c-Jun/AP-1 cascade (Nakayama et al., 2001). The MG132-augmented JNK and c-Jun phosphorylation was abolished by the addition of the JNK-c-Jun inhibitor SP600125 (10 μM) (Fig. 5).
Effects of PDE Inhibitors on TNF-α-Induced CCL2/MCP-1 mRNA and Protein Expression.Fig. 6A shows that pretreatment of VSMCs with PTX (0.5–2 mM) and cilostamide (10–40 μM) but not denbufylline (10–40 μM), dose dependently attenuated TNF-α-stimulated CCL2/MCP-1 mRNA expression. Consistent with that finding, cilostamide (40 μM) and PTX (2 mM), but not denbufylline (40 μM), attenuated TNF-α-stimulated CCL2/MCP-1 protein production (Fig. 6B).
To explore the underlying mechanisms whereby PTX and cilostamide act, VSMCs were pretreated with PTX (2 mM) or cilostamide (40 μM), followed by TNF-α (5 ng/ml) stimulation for various time points. The results showed that both PTX and cilostamide partially prevented TNF-α-induced degradation of I-κBα and nuclear translocation of p65/NF-κB (Fig. 7, A, B, and G). Additionally, PTX and cilostamide inhibited TNF-α-stimulated c-fos and c-jun mRNA expression (Fig. 7, D and E), attenuated TNF-α-activated p42/44 MAPK phosphorylation and c-Fos up-regulation, and suppressed TNF-α-induced JNK2/3 and c-Jun phosphorylation (Fig. 7, A and B). By contrast, the pretreatment of VSMCs with denbufylline, although attenuating TNF-α-activated p42/44 MAPK phosphorylation, did not affect the other TNF-α-dependent signaling pathways or TNF-α-induced c-fos and c-jun mRNA expression (Fig. 7, C, F, and G).
Discussion
The present study demonstrates that cultured rat VSMCs at steady-state expressed a very low level of CCL2/MCP-1 mRNA and protein, which could be up-regulated substantially by TNF-α in both time- and dose-dependent manners. This finding is consistent with previous reports that show CCL2/MCP-1 induction by proinflammatory cytokines in a variety of nonimmune tissue cells, including VSMCs (Hurwitz et al., 1995; Parry et al., 1998; Ortego et al., 1999; de Keulenaer et al., 2000; Iseki et al., 2000; Chen et al., 2001; Momoi et al., 2001; Loghmani et al., 2002).
The intracellular signaling pathways leading to TNF-α-dependent CCL2/MCP-1 expression may include those that activate protein kinases such as p42/44 MAPK, p38 MAPK, and JNK (Leong and Karsan, 2000). This study shows that TNF-α-stimulated CCL2/MCP-1 mRNA and protein expression was attenuated by pharmacological inhibitors of p42/44 MAPK kinase (PD98059 or U0126), but not p38 MAPK (SB203580), suggesting that activation of the p42/44 MAPK pathway is involved in TNF-α-stimulated CCL2/MCP-1 expression in rat VSMCs in culture. This finding is consistent with that reported by de Keulenaer et al. (2000). However, in other rat cell types or in human cell systems, p38 MAPK instead of p42/44 MAPK was reported to mediate the induction of CCL2/MCP-1 elicited by TNF-α (Goebeler et al., 1999; Rovin et al., 1999; Blinman et al., 2000). Thus, the MAPK signals mediating TNF-α-dependent CCL2/MCP-1 expression may vary by cell types and species. Besides p42/44 and p38 MAPKs, the present study shows that SP600125, a potent JNK inhibitor, attenuated TNF-α-induced CCL2/MCP-1 expression. This suggests that the JNK pathway may also participate in TNF-α-dependent CCL2/MCP-1 expression in VSMCs. Because PD98059 and SP600125 inhibited distinct TNF-α-dependent MAPK pathways, we surmise that both the p42/44 MAPK and the JNK pathways coordinately modulate TNF-α-induced CCL2/MCP-1 expression in rat VSMCs. In support for this notion, the pretreatment of VSMCs with PD98059 and SP600125 in combination resulted in additive inhibition of TNF-α-induced CCL2/MCP-1 expression than either agent alone.
Many of the cellular responses induced by TNF-α require intracellular signals transduced to two major transcription factors, NF-κB and AP-1 (Karin, 1995; Kyriakis, 1999; Leong and Karsan, 2000; Baud and Karin, 2001). Upon stimulation with TNF-α, activated p42/44 MAPK and JNK can enter the nucleus to induce expression of c-Fos and c-Jun proteins, respectively. These proteins then combine to form c-Jun-based AP-1 complexes. In addition, the JNKs, especially JNK2 isoform, can efficiently phosphorylate the N-terminal sites of c-Jun and enhance AP-1 transcriptional activity (Kallunki et al., 1994). In this study, we show that pretreatment with SP600125 completely inhibited TNF-α-activated phosphorylation of JNK2/3 and c-Jun. This is consistent with our current understanding that SP600125 is a potent inhibitor for the JNK-c-Jun pathway (Bennett et al., 2001). By contrast, PD98059 was found to abolish TNF-α-induced p42/44 MAPK phosphorylation and the resultant c-Fos up-regulation. Because neither SP600125 nor PD98059 affected TNF-α-induced p65/NF-κB nuclear translocation, their anti-CCL2/MCP-1 action might be mediated predominantly through down-regulation of AP-1 activity. However, AP-1 proteins are known to interact with NF-κB and modulate κB-dependent gene transcription (Kyriakis, 1999), the possibility that SP600125 and PD98059 may act in part through attenuation of nuclear NF-κB activity cannot be ruled out.
Activation of NF-κB by TNF-α involves phosphorylation-dependent ubiquitination and degradation of I-κB proteins, which normally trap NF-κB within the cytoplasm of quiescent cells (Chen and Goeddel, 2002). Freed NF-κB can enter the nucleus, which in turn regulates the transactivation of κB-dependent genes involved in a variety of inflammatory disorders, including atherosclerosis (Barnes and Karin, 1997; Bourcier et al., 1997). This study shows that the ubiquitin/proteosome inhibitor MG132 attenuated but did not abolish TNF-α-stimulated CCL2/MCP-1 production. The incomplete inhibition might be explained in part by the realization that MG132 also activates the c-Jun/AP-1 pathway, which by itself could lead to transcriptional induction of CCL2/MCP-1 (Nakayama et al., 2001). When the activated c-Jun/AP-1 signal was blocked by simultaneous pretreatment of SP600125, MG132 completely inhibited TNF-α-stimulated CCL2/MCP-1 induction. Together, these findings indicate that TNF-α-induced CCL2/MCP-1 production in rat VSMCs is dually regulated by both AP-1 and NF-κB pathways.
This study shows that PTX, a nonselective PDE inhibitor, suppressed TNF-α-induced CCL2/MCP-1 production that was preceded by attenuation of TNF-α-activated phosphorylation of p42/44 MAPK, JNK2/3, and c-Jun; inhibition of TNF-α-up-regulated c-Fos expression; as well as attenuation of TNF-α-induced degradation of I-κBα and nuclear translocation of p65/NF-κB. Thus, the anti-CCL2/MCP-1 activity of PTX seems to be mediated by its ability to counteract both AP-1 and NF-κB activation triggered by TNF-α. This study further demonstrates that the selective type III PDE inhibitor cilostamide, but not the type IV inhibitor denbufylline, attenuated TNF-α-stimulated CCL2/MCP-1 induction and down-regulated TNF-α-dependent signaling pathways similarly to that achieved by PTX treatment. This suggests that type III PDE may be a major, if not the only, target whereby PTX acts to suppress TNF-α-stimulated CCL2/MCP-1 induction. Our result was consistent in part with that observed by Aizawa et al. (2003) who showed that inhibition of PDE3 by cilostamide activated PKA and dose dependently decreased TNF-α-induced NF-κB-dependent reporter gene expression in VSMCs. In view of the critical role of CCL2/MCP-1 in atherogenesis, the ability of PTX and cilostamide to antagonize cytokine-induced CCL2/MCP-1 expression may have implications in the prevention and treatment of atherosclerotic vascular disorders.
In summary, TNF-α-stimulated CCL2/MCP-1 induction in rat VSMCs is dually regulated by AP-1 and NF-κB pathways. Inhibition of type III PDE may contribute substantially to the suppressive effect of PTX on TNF-α-dependent CCL2/MCP-1 production via down-regulation of AP-1 and NF-κB signals.
Acknowledgments
We thank Yi-Hsuan Lee for excellent technical assistance.
Footnotes
-
This work was supported by grants from the National Taiwan University Hospital 93S027 (to Y.-M.C.), the Ta-Tung Kidney Foundation, and the Mrs. Hsiu-Chin Lee Kidney Research Fund, Taipei, Taiwan.
-
DOI: 10.1124/jpet.103.062620.
-
ABBREVIATIONS: CCL2, CC chemokine ligand 2; MCP-1, monocyte chemoattractant protein-1; CCR2, chemokine receptor 2; VSMC, vascular smooth muscle cell; TNF, tumor necrosis factor; NF-κB, nuclear factor-κB; AP-1, activator protein-1; I-κB, inhibitory protein of NF-κB; MAPK, mitogen-activated protein kinase; JNK, c-Jun NH2-terminal kinase; PTX, pentoxifylline; PDE, phosphodiesterase; DMEM, Dulbecco's modified Eagle's media; FCS, fetal calf serum; PD98059, 2-(2-amino-3-methoxyphenyl)-4H-1-benzopyran-4-one; U0126, 1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio) butadiene; SB203580, 4-[5-(4-fluorophenyl)-2-[4-(methylsulfonyl)phenyl]-1H-imidazol-4-yl]pyridine; MG132, carbobenzoxy-l-leucyl-l-leucyl-l-leucinal; SP600125, anthra[1-9-cd]pyrazol-6(2H)-one; PBS, phosphate-buffered saline; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
- Received November 6, 2003.
- Accepted February 19, 2004.
- The American Society for Pharmacology and Experimental Therapeutics