Inhibition of Farnesyltransferase Prevents Collagen-Induced Arthritis by Down-Regulation of Inflammatory Gene Expression through Suppression of p21^ras-Dependent NF-κB Activation¹

Inflammatory processes contribute to the pathologic events that lead to tissue injury and destruction in autoimmune diseases and other inflammatory conditions. Rheumatoid arthritis (RA)³ is a typical immune-mediated disease characterized by chronic inflammation in the synovium, the formation of pannus tissue, and destruction of joint tissues (1). Production of IL-1 and TNF-α plays a central role in the pathogenesis of RA (2). These cytokines induce the expression of inducible NO synthase (iNOS) and cyclooxygenase-2 (COX-2), resulting in the production of PGE₂ and NO. Up-regulation of the expression of vascular adhesion molecules then leads to the infiltration of various inflammatory cells into the synovial tissues of patients with RA (3). Therefore, inhibition of the biological activities of TNF-α and IL-1 by their neutralizing Abs or receptor antagonists can be an effective therapy for inflammatory arthritis (4, 5). Inhibition of iNOS and COX-2 activities by N-monomethyl-l-arginine and celecoxib, respectively, suppresses the development of arthritis (6, 7).

Expression of these cytokines and enzymes can be regulated by the activation of the transcription factor NF-κB, which is involved in several aspects of the pathogenesis of RA and other chronic inflammatory diseases (8, 9). NF-κB is activated as a consequence of phosphorylation, ubiquitination, and subsequent proteolytic degradation of I-κB through activation of the I-κB kinase (IKK) (10). The liberated NF-κB translocates into nuclei and binds as a transcription factor to κB motifs in the promoter of target genes, leading to the induction of their mRNA expression. Therefore, regulation of NF-κB activation can be an extremely attractive target for therapeutic intervention of inflammatory processes and RA development (8, 9). Many anti-inflammatory drugs, such as glucocorticoids, nonsteroidal anti-inflammatory drugs, and immunosuppressants, act as inhibitors of the NF-κB pathway and suppress expression of various inflammation-associated genes, such as iNOS, COX-2, IL-1β, and TNF-α. Therefore, inhibitors of the NF-κB pathway have been used for treatment of inflammatory diseases, including arthritis (11).

Farnesyltransferase (FTase) inhibitor (FTI) suppresses the production of IL-1β and NO (12, 13), thus indicating that the farnesylation of p21^ras may regulate the intracellular signal pathway in inflammatory gene expression. The activation of membrane-bound p21^ras increases the kinase activity of the cytosolic serine/threonine kinase Raf-1 (14) and PI3K (15), which result in the enhancement of transcriptional activities of different transcription factors, including the NF-κB/Rel family (15, 16). These data indicate that Ras-Raf and Ras-PI3K pathways may increase the expression of NF-κB-dependent inflammatory genes by activating a sequential cascade of IKK-dependent I-κBα phosphorylation and degradation and nuclear translocation of cytosolic NF-κB (17). We, therefore, hypothesized that inhibition of FTase can suppress the production of inflammatory gene products by blocking p21^ras-dependent NF-κB pathway. In the present study, we synthesized the nonpeptide pyrrole-based FTI LB42708 with high potency and selectivity and investigated the effects of LB42708 on the production of inflammatory mediators and the development of collagen-induced arthritis (CIA). Our data showed that LB42708 suppressed NF-κB-dependent inflammatory cytokine expression and production of NO and PGE₂ by suppressing IKK activity. Furthermore, this compound significantly attenuated the arthritic incidence and severity in a mouse CIA model, suggesting that LB42708 may be a beneficial agent for treating chronic inflammatory arthritis.

RAW264.7 cells (murine macrophage cell line) and ROS17/2.8 cells (rat osteoblast-like osteosarcoma cells) obtained from American Type Culture Collection (Manassas, VA) were cultured in DMEM (Invitrogen Life Technologies, Rockville, MD) containing 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% FBS in a humidified incubator with 5% CO₂/95% air at 37°C. RAW264.7 cells and ROS17/2.8 cells were pretreated with LB42708 (R&D Center, LG Life Sciences, Daejon, Korea) or SCH66336 (Schering-Plough Research Institute, Kenilworth, NJ) for 12 h, and these cells were stimulated with LPS (1 μg/ml) + IFN-γ (100 U/ml) and cytokine mixture (CM; 10 U/ml IL-1β, 100 U/ml IFN-γ, and 25 ng/ml TNF-α), respectively. Peritoneal macrophages were harvested from female BALB/c mice (6–8 wk old; Daehan Biolink, Daejeon, Korea) at day 5 following i.p. injection with 1.5 ml of sterile 4% thioglycolate broth (Difco Laboratories, Detroit, MI) and were cultured in 96-well plates (2 × 10⁵ cells/well) or 6-well plates (1 × 10⁷ cells/well) at 37°C in 5% CO₂/95% air for 24 h. The cells were then stimulated with LPS + IFN-γ, following pretreatment with FTI for 12 h.

Male DBA/1J mice and BALB/c mice (6–8 wk old) were obtained from Daehan Biolink and maintained at the specific pathogen-free housing facility of the School of Medicine, University of Kangwon National University (Chunchon, Korea). All procedures performed on these animals were in accordance with the guidelines of the University Animal Care and Use Committee. Mice were immunized intradermally at the base of the tail with 100 μg of type II bovine collagen (Chondrex, Seattle, WA) emulsified in CFA (n = 14) or saline (n = 7). At day 15, immunized mice were divided into two groups; group A (n = 7) was i.p. injected with saline and group B (n = 7) was i.p. injected with LB42708 (10 mg/kg) daily. Twenty-one days after primary immunization, mice were boosted with 100 μg of type II bovine collagen in IFA. To accelerate the development of arthritis (18), 20 μg of LPS (Escherichia coli, serotype 055:B5; Sigma-Aldrich, St. Louis, MO) in sterile saline was i.p. injected in mice on day 28. Mice were examined daily for the onset of CIA. The swelling of four paws was graded from 0 to 4, as follows: grade 0, no swelling; grade 1, swelling of finger joints or focal redness; grade 2, mild swelling of wrist or ankle joints; grade 3, severe swelling of the entire paw; and grade 4, deformity or ankylosis. Each paw was graded, and the four scores were totaled, so that the maximal score per mouse was 16. Incidence was expressed as the number of mice that showed paw swelling among the total number of mice examined. The time of onset was expressed as the mean time when paw swelling was first observed in individual mice. For some experiments, BALB/c mice were i.p. injected with LB42708 (12.5 mg/kg), and 4 h later injected with LPS (2 mg/kg). After 12 h of LPS injection, blood samples were collected by cardiac puncture, and serum was prepared by centrifugation at 12,000 × g for 30 min.

NO production was determined by measuring the amount of nitrite in the culture medium and nitrite plus nitrate (NOx) in serum using Griess reagents (19) and nitrate reductase-based colorimetric assay kit (Alexis, San Diego, CA). The levels of PGE₂, TNF-α, and IL-1β were determined using ELISA kits purchased from Amersham Biosciences (Piscataway, NJ) and R&D Systems (Minneapolis, MN).

Western blot analysis was performed, as previously described (18). In brief, cells were harvested, washed twice with ice-cold PBS, suspended in 10 mM Tris-HCl (pH 7.4), and lysed by three cycles of freezing and thawing. Cell extracts were obtained by centrifugation at 12,000 × g at 4°C for 20 min. Cytosolic proteins (30 μg) were electrophoretically resolved on 12% SDS-PAGE and transferred onto nitrocellulose membranes. The membranes were blocked with 5% nonfat dried milk and hybridized with iNOS mAb (BD Transduction Laboratories, Lexington, KY) and polyclonal COX-2 Ab (Santa Cruz Biotechnology, Santa Cruz, CA). After washing four times, the membranes were hybridized with HRP-conjugated secondary Abs. The membranes were incubated for 2 min with ECL solution and exposed to x-ray film. For determining Ras processing, whole cell lysates (50 μg) were resolved on 12.5% SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was immunoblotted with anti-Ras Ab (Calbiochem, Cambridge, MA) and peroxidase-conjugated secondary Ab.

Cells were harvested and washed with PBS, and the pellets were resuspended in 80 μl of immunoprecipitation lysis buffer (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, 10% glycerol, 1% Nonidet P-40, 5 mM EDTA, 1 mM DTT, 100 mM NaF, 2 mM sodium pyrophosphate, 2 mM sodium orthovanadate, 1 mM PMSF, 10 μg/ml aprotinin, and 10 μg/ml leupeptin) and stored on ice for 20 min before centrifugation (14,000 × g, 20 min, 4°C). IKK complex was immunoprecipitated by incubation for 1 h at 4°C with polyclonal IKKα Ab (from Santa Cruz Biotechnology) bound to protein A-Sepharose. The immunoprecipitates were washed twice with immunoprecipitation buffer and twice with kinase buffer (20 mM HEPES, pH 7.4, 20 mM β-glycerophosphate, 20 mM MgCl₂, 2 mM DTT, and 0.1 mM sodium orthovanadate). The kinase assays were initiated by the addition of 2 μM GST-I-κBα fusion protein as substrate and 0.5 μCi of [γ-³²P]ATP. Reaction mixtures were incubated for 30 min at 30°C and stopped by the addition of 2× SDS-PAGE sample buffer. The phosphorylation of the I-κBα proteins was examined by SDS-PAGE, followed by autoradiography and densitometry.

Cells were pretreated with FTI for 12 h and treated with LPS (1 μg/ml) + IFN-γ (100 U/ml) for 2 h. Nuclear extracts were prepared, as described previously (19). Briefly, cells were washed and scraped into phosphate-buffered solution, and centrifuged at 4,500 rpm for 5 min in a microfuge (Beckman Coulter, Fullerton, CA). The pelleted cells were suspended in buffer A (10 mM Tris, pH 7.5, 1.5 mM MgCl₂, 10 mM KCl, 0.5% Nonidet P-40) at ∼10 times the packed cell volume and lysed by gentle pipetting. Nuclei were recovered by microcentrifugation at 7,000 rpm for 5 min. The nuclear proteins were extracted at 4°C by gentle resuspension of the nuclei (at approximately twice the packed nuclear volume) in Tris buffer (20 mM Tris (pH 7.5), 10% glycerol, 1.5 mM MgC1₂, 420 mM NaCl, and 0.2 mM EDTA), followed by 30 min of platform rotation. The nuclear protein suspension was obtained by centrifugation at 12,000 × g for 15 min. All buffers contained the following additions: 1–2 μg/ml each of aprotinin, chymostatin, leupeptin, and pepstatin; 0.2 mM PMSF; 0.5 mM DTT; and 0.1 mM sodium vanadate. All steps were conducted on ice or at 4°C. A double-stranded NF-κB-specific probe (5′-AGTTGAGGGGACTTTCCCAGGC-3′; Promega, Madison, WI) was labeled with [γ-³²P]ATP using T4 polynucleotide kinase and purified on a G-50 Sephadex column. The ³²P-labeled probe (∼40,000 cpm) was then incubated with the nuclear extracts (10 μg of protein) for 20 min at room temperature. Samples were resolved on native 5% polyacrylamide gel, and the gel was dried and subjected to autoradiography.

DNA transfections of cells were conducted in six-well plates by using lipofectamine (Invitrogen Life Technologies), as described (19). Briefly, 1 μg of a murine iNOS promoter-luciferase construct was incubated with 20 μg of lipofectamine at room temperature for 20 min and diluted in 1 ml of serum-free medium. Cells were exposed to this mixture solution in a humidified incubator with 5% CO₂/95% air at 37°C for 4 h, washed, and cultured overnight in DMEM supplemented with 10% FBS. The cells were washed with fresh medium, pretreated with LB42708 for 12 h, and stimulated with LPS for 12 h. Subsequently, cells were lysed with buffer containing 1% Triton X-100, 5 mM DTT, 50% glycerol, 10 mM EDTA, and 125 mM Tris-phosphate (pH 7.8), and luciferase activity was measured by luminometer.

Synovial tissues were isolated from the joints of hind paws at day 30 after the first immunization, and total RNA was extracted using a TRIzol reagent kit (Invitrogen Life Technologies). A total of 3 μg of mRNA was converted to cDNA by treatment with 200 U of reverse transcriptase and 500 ng of oligo(dT) primer in 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl₂, 10 mM DTT, and 1 mM dNTPs at 42°C for 1 h. The reaction was stopped by heating at 70°C for 15 min. A total of 3 μl of the cDNA mixture was used for enzymatic amplification. PCR was performed in 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl₂, 0.2 mM dNTPs, 2.5 U of TaqDNA polymerase, and 0.1 μM of each primer for iNOS, COX-2, TNF-α, and IL-1β. The amplification was performed in a DNA thermal cycler under the following condition: denaturation at 94°C for 5 min for the first cycle and for 45 s starting from the second cycle; annealing of iNOS at 47°C for 45 s; annealing of COX-2, TNF-α, and IL-1β at 51°C for 45 s; and extension at 72°C for 30 s for 35 cycles. Final extension was performed at 72°C for 10 min. The PCR products were electrophoresed on a 1.5% agarose gel and stained with ethidium bromide. The primers used were 5′-TTTGGAGCAGAAGTGCAAAGTCTC-3′ (sense) and 5′-GATCAGGAGGGATTTCAAAGACCT-3′ (antisense) for the iNOS; 5′-CCGTGGTGAATGTATGAGCA-3′ (sense) and 5′-CCTCGCTTCTGATCTGTCTT-3′ (antisense) for the COX-2; 5′-ATGAGCACAGAAAGCATG-3′ (sense) and 5′-TCACAGAGCAATGACTCC-3′ (antisense) for the TNF-α; 5′-ATGGCAACTGTTCCTGAAC-3′ (sense) and 5′-TTAGGAAGACACGGATTC-3′ (antisense) for the IL-1β; and 5′-TCCTTCGTTGCCGGTCCA CA-3′ (sense) and 5′-CGTCTCCGGAGTCCATCACA-3′ (antisense) for the β-actin. The PCR products were electrophoresed on 1.5% agarose gel and stained with ethidium bromide to visualize DNA. The image was acquired using an image analyzer.

The data are presented as means ± SD of at least three separate experiments. Comparisons between two groups were analyzed using Student’s t test, and significance was established at a p value <0.05.

We recently showed that the nonpeptidic FTI LB42908 potently and selectively inhibited the farnesylation of Ras proteins (20). To investigate the anti-inflammatory mechanism of FTI, we first attempted to improve the potency and/or selectivity of our first generation of LB42908 and synthesized the nonpeptide pyrrole-based LB42708 (Fig. 1,A), which inhibited FTase activities toward H-ras, N-ras, and K-ras4B with IC₅₀ values of 0.8, 1.2, and 2.0 nM, respectively, and geranylgeranyltransferase I activity with an IC₅₀ of 100 μM. This inhibitor was highly selective (>50,000-fold) for FTase over the closely related geranylgeranyltransferase I (IC₅₀ = 100,000 nM), and this selectivity was superior to those of other commercial FTIs such as FTI-276 (21), R115777 (22), SCH66336 (23), and LB42908 (20). We next investigated the inhibitory effect of LB42708 on the processing of the cellular farnesylated protein p21^ras. Stimulation of murine macrophage cell line RAW264.7 cells with LPS + IFN-γ increased p21^ras processing compared with control cells; however, this increased processing was inhibited by pretreatment with LB42708 in a concentration-dependent manner (with an IC₅₀ of ∼10 nM) (Fig. 1 B). The inhibitory effect of LB42708 was stronger than that of the commercial FTI SCH66336 at 50 nM. These results demonstrate that LB42708 is a potent and selective inhibitor of FTase, and is therefore efficacious in inhibiting the cellular processing of p21^ras.

It has been shown that the activation of membrane-bound Ras increases the kinase activity of a cytosolic serine/threonine kinase Raf-1 (14), which results in the enhancement of transcriptional activity of NF-κB (16). NF-κB activation requires a sequential cascade such as IKK-dependent I-κBα phosphorylation, ubiquitination and degradation, translocation of cytosolic NF-κB to nucleus, and binding to its consensus sequence of several gene promoters (17, 24, 25). We first measured the effect of LB42708 on IKK activity and I-κBα phosphorylation in LPS-stimulated RAW264.7 cells. When RAW264.7 cells were stimulated with LPS, I-κBα phosphorylation and cytosolic IKK activity were significantly increased; however, these increases were suppressed by LB42708 (Fig. 2, A and B). We next examined the effects of LB42708 on the proteolytic degradation of I-κBα. Western blot analysis showed that LPS treatment significantly decreased the I-κBα protein level compared with the control, and this decrease was blocked by LB42708 treatment (Fig. 2,C). To further examine the effects of FTI on NF-κB translocation to nucleus, we determined the cytosolic and nuclear NF-κB p65 subunit levels following treatment of RAW264.7 cells with LPS in the presence or absence of LB42708 by Western blot analysis. As shown in Fig. 2,D, a decrease in the cytosolic p65 subunit level following LPS treatment was accompanied with an increase in the nuclear p65 level, whereas the subunit was mostly present in the cytosol of the control cells. However, LB42708 treatment inhibited a LPS-induced increase in translocation of the cytosolic p65 subunit to the nuclei. We next examined the effects of LB42708 on NF-κB-binding activity to its consensus oligonucleotide by electromobility gel shift assay. When RAW264.7 cells were stimulated with LPS, protein-DNA-binding activity in the nuclear extracts was increased compared with control (Fig. 2,E). However, this binding activity was significantly inhibited by pretreatment with LB42708. Specificity of the DNA-protein interaction for NF-κB was demonstrated by competition assay with cold probe, but not with cold mutant probe. A supershift study with specific Ab against NF-κB p65 demonstrated the presence of NF-κB in the complex. NF-κB activation is important for expression of inflammatory genes, including iNOS, which contains a functional NF-κB binding site in its promoter (11). Thus, we next examined the effect of LB42708 on iNOS promoter activity. RAW264.7 cells transfected with an iNOS promoter construct increased the promoter activity by ∼3-fold in response to LPS, and this increase was suppressed by LB42708 in a dose-dependent manner (Fig. 2 F). These results indicate that farnesylation of p21^ras is involved in intracellular signaling cascade for NF-κB activation through IKK-dependent phosphorylation and degradation of I-κBα.

Because macrophages have been shown to play a pivotal role in the etiology of immune diseases, such as RA by producing numerous inflammatory mediators including NO and PGE₂ (6, 9, 26, 27, 28), we examined whether LB42708 could inhibit the production of NO and PGE₂ in macrophages stimulated with LPS + IFN-γ. Stimulation of RAW264.7 cells with LPS + IFN-γ increased the concentrations of nitrite and PGE₂ in the culture medium, and these increases were inhibited by pretreatment with LB42708 (Fig. 3, A and B). Its inhibitory effects were significantly higher than those of SCH66336. We next investigated whether LB42708 could inhibit the expression of iNOS and COX-2 in RAW264.7 cells. Western blot analyses showed that stimulation of RAW264.7 cells with LPS + IFN-γ increased the protein levels of iNOS and COX-2 compared with those of control cells (Fig. 3,C). These increases were suppressed by treatment with LB42708 in a concentration-dependent manner, but not effectively by SCH66336. Because osteoblasts also produce NO and PGE₂, which can induce cartilage destruction in RA (29), we examined the effect of LB42708 on the production of NO and PGE₂ and the expression of iNOS and COX-2 in rat osteoblast cell line ROS17/2.8 following stimulation with CM for 36 h. The stimulated ROS17/2.8 cells increased the concentrations of nitrite and PGE₂ in the culture medium, and these increases were significantly suppressed by pretreatment with LB42708, but not effectively by SCH66336 (Fig. 3, D and E). We further examined the effect of LB42708 on the expression of iNOS and COX-2 in ROS17/2.8 cells. When treated with CM, the protein levels of iNOS and COX-2 were significantly increased compared with those of control cells (Fig. 3 F). This induction was effectively inhibited by pretreatment with LB42708, but not by SCH66336.

Because RAW264.7 cells are a well-established macrophage cell line, we examined anti-inflammatory effects of LB42708 in primary cultured peritoneal macrophages. Peritoneal macrophages were therefore stimulated with LPS + IFN-γ for 24 h, and the levels of NO, PGE₂, TNF-α, and IL-1β were measured in the culture medium. Increased levels of nitrite and PGE₂ were observed in the peritoneal macrophages in response to LPS + IFN-γ, and these increases were reduced by pretreatment with LB42708 (Fig. 4, A and B). Similarly, treatment of peritoneal macrophages with LPS + IFN-γ increased the levels of TNF-α and IL-1β in the culture medium, and the elevated cytokine levels were markedly inhibited by pretreatment with LB42708 (Fig. 4, C and D). The suppressive effects of LB42708 on the production of these inflammatory cytokines and mediators were higher than those of SCH66336.

We examined the effect of LB42708 on the in vivo production of NO, PGE₂, TNF-α, and IL-1. LPS-administrated mice were found to have increased serum level of NOx compared with saline-treated control, and this increase was inhibited by i.p. preadministration of LB42708 (Fig. 5,A). In addition, treatment of mice with LB42708 significantly reduced the inflammatory response to LPS, as measured by serum levels of PGE₂, TNF-α, and IL-1β (Fig. 5, B–D). These results indicate that LB42708 possesses an anti-inflammatory potential by inhibiting the in vivo production of these proinflammatory mediators under pathological conditions.

To determine the effect of LB42708 on the development of CIA, DBA/1 mice were immunized with collagen and monitored for the appearance of clinical signs of arthritis. DBA/1 mice gradually developed an onset of CIA at day 28 after the first immunization of collagen, and the incidence of arthritis reached 100% by day 36 (Fig. 6,A). However, this development was reduced to 50% by administration of LB42708 from day 15 after the first immunization. Furthermore, LB42708 administration decreased arthritic severity in CIA mice (Fig. 6,B). Fig. 6,C shows that collagen injection induced arthritic swelling in the hind paws. This inflammatory swelling was attenuated by administration of LB42708. We next examined the mRNA levels of proinflammatory cytokines and enzymes in the synovium of CIA mice. Consistent with previous reports (1, 2, 6), the high levels of TNF-α, IL-1β, iNOS, and COX-2 mRNAs were observed in the synovial tissues of CIA mice, but administration of LB42708 significantly inhibited the expression of these inflammatory genes (Fig. 6 D). These results suggest that LB42708 treatment reduced the arthritic manifestations by down-regulating the expression of proinflammatory cytokines and enzymes in the joints.

This study was undertaken to elucidate the potential effect and molecular mechanism of LB42708, a nonpeptidic, highly selective FTI, on proinflammatory cytokines and mediators in vitro and in vivo and to determine its antiarthritic activity in CIA mice, a surrogate model for human RA. We found that by suppressing NF-κB activation, LB42708 inhibited the expression of iNOS, COX-2, TNF-α, and IL-1β as well as the production of NO and PGE₂ in immune-activated macrophages and osteoblasts in vitro and LPS-administrated mice. Moreover, LB42708 ameliorated the pathogenic development of CIA in DBA/1 mice and suppressed inflammatory gene expression in paws of CIA mice. These data indicate that the farnesylation of Ras proteins is a critical step in the intracellular signaling cascade for the process of inflammation, and that LB42708 possesses a potent antiarthritic activity, probably by suppressing NF-κB-dependent inflammatory gene expression.

The well-known immune stimulant LPS induces several intracellular signaling pathways in monocytes/macrophages. LPS binding to CD14 and TLR, which are both the major LPS receptors on monocytes (30), results in the activation and farnesylated modification of Ras (31, 32). The activation of Ras increases the activities of Raf-1 and PI3K (15), which are the downstream signaling components of Ras. Ras-mediated gene expression is induced by NF-κB activation and transcriptional activities through both Raf and PI3K pathways. The Ras-Raf pathway appeared to activate NF-κB through MAPK-dependent IKK2 activation, whereas the Ras-PI3K pathway activated NF-κB via Akt-dependent IKK1 activation. Therefore, the activation of Ras is able to elevate the IKK1/2 activity and lead to I-κBα degradation, thus resulting in the NF-κB-dependent inflammatory gene expression (15). Our data showed that LB42708, a selective FTI, suppressed LPS-mediated increase in p21^ras farnesylation in RAW264.7 cells (Fig. 1,B) and effectively inhibited in vitro and in vivo the expression of inflammatory genes such as iNOS, COX-2, TNF-α, and IL-1β (Figs. 3 and 5). Moreover, the inhibitors of Raf-1 (Bay43-9006) and PI3K (LY294002), but not MEK inhibitor (PD98059), effectively suppressed iNOS expression, NO production, and iNOS promoter activity, which are regulated by NF-κB, in LPS-stimulated RAW264.7 cells (data not shown). These results indicate that Raf-1, PI3K, and Akt are the signaling components linked between Ras and NF-κB-dependent inflammatory gene expression in LPS-stimulated macrophages. This evidence suggests that p21^ras fanesylation and activation may be apical events in the intracellular signaling pathway of inflammatory gene expression in LPS-activated macrophages. However, other possible mechanisms by which FTI may inhibit NF-κB activation by suppressing IL-1R-associated kinase 1/TNFR-associated factor 6 complex formation (33) and reactive oxygen generation (34) cannot be completely excluded.

NO and PGE₂ are two important mediators of inflammatory joint disease. They are produced by iNOS and COX-2 expression, respectively, within macrophages that are highly prevalent in the inflamed synovial membrane and at the cartilage-pannus junction (6, 7). Many studies showed that the levels of nitrate and PGE₂ in serum or synovial fluid are significantly higher in adjuvant-induced arthritic animals as well as patients with RA than healthy controls (35, 36). Treatment of mice with iNOS or COX-2 inhibitor ameliorated pathogenesis or development of inflammatory arthritis (6, 7), and iNOS gene-deleted mice showed delayed development of CIA and decrease in arthritis severity (37). Consistent with previous reports that FTI could inhibit NO production and iNOS expression in vascular smooth muscle cells (12), our data showed that LB42708 suppressed the production of NO and PGE₂ in immune-activated macrophages and osteoblasts in vitro as well as LPS-administrated mice. These suppressive effects were found to be directly associated with down-regulation of iNOS and COX-2 protein expression. NO highly reacts with superoxide to produce the strong oxidant peroxynitrite, which in turn interacts with tyrosine residues on a variety of proteins to form nitrotyrosine found in inflamed joints of CIA mice (37). Peroxynitrite has also been implicated in the activation of COX to produce PGs (38) as well as inhibition of the antioxidant enzymes, superoxide dismutase and catalase, to increase oxidative stress (39, 40). Furthermore, FTI has been shown to inhibit superoxide production in vascular smooth muscle cells (41). This evidence indicates that FTI may suppress oxidative stress through the inhibition of NO and superoxide production, both of which are involved in the pathogenesis of inflammatory arthritis (37). These data suggest that LB42708 can inhibit the production of inflammatory mediators by suppressing inflammatory signal pathway and prevent oxidative injury in vitro and in vivo.

TNF-α and IL-1β, proinflammatory cytokines, are produced by immune-activated macrophages and mediate the pathogenesis and progression of RA (1, 9). Administration of neutralizing Abs for TNF-α or IL-1β significantly reduces the clinical and histological severity of this disease (4, 42). Moreover, comparative studies of murine CIA have shown that administration of anti-TNF-α Ab was effective in slowing the onset of arthritis, whereas anti-IL-1β Ab was effective in reducing cartilage destruction (42, 43). These studies indicate that TNF-α and IL-1β play key roles in development and progression of inflammatory arthritis. In the present study, we showed that LB42708 inhibited the production of these cytokines in immune-activated macrophages in vitro and LPS-administrated mice. Moreover, this compound markedly decreased the levels of mRNA for TNF-α, IL-1β, iNOS, and COX-2 in the inflamed joints of CIA mice. These data indicate that inhibition of FTase blocks inflammation-associated gene expression at transcriptional level. This anti-inflammatory effect may be directly responsible for the amelioration of incidence and severity of arthritis in collagen-treated mice.

NF-κB is a common transcription factor for regulating the expression of many inflammatory genes, including iNOS, COX-2, TNF-α, and IL-1β, by interacting with its DNA-binding motif on the promoters of these genes (1, 9, 11). Thus, the abnormal, constitutive activation of NF-κB has been associated with a number of chronic inflammatory arthritides (9). Many antiarthritic drugs, including nonsteroidal anti-inflammatory drugs, inhibit the expression of proinflammatory genes, such as iNOS, COX-2, TNF-α, and IL-1β, through suppression of the NF-κB pathway (27). This evidence suggests that the regulation of NF-κB activation may be an attractive target for the prevention or treatment for arthritic diseases. This transcription factor can be activated by Raf and/or PI3K-dependent IKK activation through the farnesylation of p21^ras (15), suggesting that FTI can inhibit the NF-κB pathway and the production of proinflammatory cytokines upstream of NF-κB. It has been shown that the inhibition of protein farnesylation blocked NO production by inhibiting iNOS expression at the transcriptional level (12). In the present study, LB42708 was found to inhibit IKK activity, I-κB degradation, and translocation of the cytosolic NF-κB p65 subunit to the nucleus, resulting in increased iNOS promoter activity and inflammatory protein expression. These results, therefore, suggest that the inhibition of p21^ras farnesylation by LB42708 suppresses NF-κB activation by blocking IKK activation, and subsequently inhibits the production of the proinflammatory cytokines and mediators.

The p21^ras has been found to be localized in synovial lining cells in ∼70% of the RA cases (44), and mutational activation of this protein has been shown to occur in the synovial tissue of patients with RA (45). These studies suggest that inhibition of p21^ras activation by FTI could ameliorate pathogenesis of RA. Indeed, our data showing that LB42708 significantly suppressed arthritic incidence and severity in CIA mice support the involvement of p21^ras activation in the pathogenesis of RA. The antiarthritic mechanism of LB42708 is most likely due to the inhibition of inflammatory gene expression through the inhibition of p21^ras farnesylation and NF-κB activation, thus providing clear evidence that LB42708 may be a potential therapeutic drug for treating RA.

We thank Elaine Por for preparing the manuscript.