Signaling Events Involved in Macrophage Chemokine Expression in Response to Monosodium Urate Crystals*

Chemokine production has been associated with leukocyte infiltration into the joint during gouty arthritis, and monosodium urate (MSU) crystals, the causative agent of this arthropathy, have been shown to modulate their expression. In the present study, we investigated the transductional mechanisms underlying this cellular regulation in the murine macrophage cell line B10R. We report that MSU crystals rapidly and transiently increase mRNA levels of various chemokines in a concen-tration-dependent manner. Examination of second mes-senger activation revealed that macrophage exposure to MSU crystals led to MEK1/2, ERK1/2, and inhibitory protein (cid:1) B (cid:2) phosphorylation as well as to NF- (cid:1) B and AP-1 nuclear translocation. Of interest, specific blockage of the ERK1/2 pathway drastically reduced up-modulation of MSU crystal-mediated Ab, antibody; EMSA, electrophoretic mobility shift assay; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MTS, (3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenil)-2-(4-sulfophenyl)-2H-tetrazolium,innersalt. ficity was tested by adding to nuclear extracts from 2-h treated macrophages a 100-fold molar excess of either a cold AP-1 consensus oligonucleotide (100 (cid:1) specific) or a nonspecific Sp1 probe (100 (cid:1) nonspe-cific). C , nuclear proteins from MSU crys- tal-stimulated cells (100 (cid:5) g/ml for 2 h) pretreated or not with 5–40 (cid:5) M apigenin or 20–40 (cid:5) M PD 98059 were subjected to EMSA following incubation with an AP-1 consensus probe. Binding specificity was tested by adding to nuclear extracts from 2-h treated cells a 100-fold molar excess of cold AP-1 oligonucleotide. These results are representative of one of three independent experiments.

Gouty arthritis is characterized by joint inflammation as a result of intra-articular deposition of monosodium urate (MSU) 1 crystals in individuals with elevated serum concentrations of uric acid (1). MSU crystals initiate, amplify, and sustain intense attacks of acute inflammation because of their ability, among others, to stimulate the release of proinflammatory mediators, leading to endothelium activation and leukocyte recruitment (2). In association with aggressive and destructive outcome, high amounts of proinflammatory cytokines (e.g. tumor necrosis factor-␣, IL-1␤, and IL-6) have been found in various arthropathies, including gout (reviewed by Punzi et al. (3)). In addition to these cytokines, accumulating evidence indicates that chemokines and myeloid-related proteins, which are powerful leukocyte chemoattractants and activators (4,5), also contribute to acute gout inflammation (6 -9). Indeed, increased levels of chemokines IL-8/CXCL8 (6) and monocyte chemoattractant protein (MCP)-1/CCL2 (7) as well as of myeloid-related proteins S100A8/A9 (8) and S100A12 (9) have been detected in synovial fluid of patients suffering from gout.
Extensive data indicate that MSU crystals induce proinflammatory mediator expression due to their capacity to trigger specific cellular signaling events. In fact, MSU crystals activate multiple second messengers, including G proteins, phospholipase C and D, Src tyrosine kinases, and mitogen-activated protein kinases (MAPK) (22)(23)(24)(25)(26)(27). Importantly, MSU crystalmediated IL-8 regulation in human monocytes has been linked to selective signaling pathway activation, involving members of the Src kinases, the extracellular signal-regulated kinases 1 and 2 (ERK1/2) MAPK as well as the binding of transcription factors NF-B and AP-1 to the IL-8 promoter (26,27). Although these studies have provided a better understanding of IL-8 regulation by MSU crystals, the mechanisms underlying the expression of other chemokines that are also produced during human and experimental gout remain unexplored.
In the present study, we demonstrate that MSU crystals rapidly and transiently increase mRNA levels of CC and CXC chemokines (MIP-1␣, MIP-1␤, MCP-1, and MIP-2) in murine B10R macrophages through transcriptional as well as posttranscriptional mechanisms of control. Characterization of the signal transduction events revealed a requirement for ERK1/2 MAPK-dependent signals to NF-B and AP-1 nuclear translocation and subsequent chemokine mRNA expression. These data, along with previous reports (26,27), indicate that activation of the ERK1/2 pathway appears to be a common mechanism through which MSU crystals activate transcription of multiple chemokines.

EXPERIMENTAL PROCEDURES
Materials-Triclinic MSU crystals were kindly provided by Drs Rinaldo de Médicis and André Lussier (Université de Sherbrooke, Sherbrooke, Québec, Canada) and prepared as previously described (28). The absence of contamination by endotoxins in the MSU preparations was confirmed by performing the Limulus amebocyte lysate test (Etoxate kit (Sigma)) as well as nitric oxide measurements in presence of Polymixin B (Sigma) (data not shown), as we previously described (29). Isotopes [␣-32 P]dCTP (3000 Ci/mmol) and [␥-32 P]dATP (3000 Ci/mmol) were purchased from ICN Pharmaceuticals Canada Ltd. (Montréal, Québec, Canada). Actinomycin D (AD) was obtained from Sigma. Apigenin and PD 98059 were purchased from Calbiochem, and BAY 11-7082 was from Biomol Research Laboratories (Plymouth Meeting, PA).
RNase Protection Assays (RPAs)-mRNA expression studies were performed using an RPA kit (Riboquant, BD Pharmingen; San Diego, CA), as we described elsewhere (32). Briefly, total RNA was isolated from stimulated cells with Trizol reagent (Invitrogen) according to the manufacturer's protocol. Multiprobe mCK-5, which contains templates for the murine chemokines RANTES (regulated on activation normal T cell expressed and secreted), MIP-1␤, MIP-1␣, MIP-2, IP-10, MCP-1, and TCA-3, and the housekeeping genes ml-32 and GAPDH, was labeled with [␣-32 P]dUTP using T7 RNA polymerase. Then 3 ϫ 10 5 cpm of labeled probe was allowed to hybridize with 10 g of total RNA for 16 h at 56°C. mRNA probe hybrids were treated with RNase A and phenol/ chloroform-extracted. Protected hybrids were resolved on a 5% dena-turing polyacrylamide sequencing gel and exposed to radiographic film overnight at Ϫ80°C. Laser densitometry was performed using an ␣ Imager 2000 digital imaging and analysis system (␣ Innotech, San Leandro, CA).
Electrophoretic Mobility Shift Assay (EMSA)-Cell stimulation was terminated by the addition of ice-cold PBS, and nuclear extracts were prepared, as we described previously (32). In brief, sedimented cells were resuspended in 400 l of cold buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 1.0 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride). After 15 min on ice, 25 l of 10% (v/v) Igepal were added, and the lysate was vortexed for 10 s and centrifuged for 30 s at 12,000 ϫ g. The supernatant was discarded, and the cell pellet was resuspended in 100 l of cold buffer B (20 mM HEPES, pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride). Cells were then rocked vigorously at 4°C for 15 min. Cellular debris were removed by centrifugation at 12,000 ϫ g for 5 min at 4°C, and the supernatant was stored at Ϫ80°C until used. EMSA was performed with 6 g of nuclear extract. Protein concentrations were determined using the commercial BCA Protein Assay Reagent (Pierce). Then nuclear extracts were incubated for 20 min at room temperature in 1.0 l of binding buffer (100 mM HEPES, pH 7.9, 40% glycerol, 10% Ficoll, 250 mM KCl, 10 mM dithiothreitol, 5 mM EDTA, 250 mM NaCl), 2 g of poly(dI-dC), and 10 g of nuclease-free bovine serum albumin (fraction V) (Sigma) containing 1.0 ng of radiolabeled double-stranded DNA oligonucleotide. Then double-stranded DNA (100 ng) was endlabeled by using [␥-32 P]dATP and T4 polynucleotide kinase (New England Biolabs). This mixture was incubated for 20 min at room temperature, and the reaction was stopped using 5 l of 0.2 M EDTA. The labeled oligonucleotide was extracted with phenol/chloroform and passed through a G-50 spin column. The double-stranded DNA oligonucleotides (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), used either as probes or as competitors, were as follows: consensus binding site for AP-1 c-Jun homodimer and Jun/Fos heterodimeric complexes, 5Ј-CGCTTGATGACTCAGCCGGAA-3Ј; consensus binding site for NF-B/c-Rel homodimeric and heterodimeric complexes, 5Ј-AGTTGAGGG-GACTTTCCCAGGC-3Ј; and consensus binding site for Sp1, 5Ј-ATTC-GATCGGGGCGGGGCGAGC-3Ј. DNA-protein complexes were resolved from free-labeled DNA by electrophoresis in native 4% (w/v) polyacrylamide gels containing 50 mM Tris-HCl, pH 8.5, 200 mM glycine, and 1 mM EDTA. The gels were subsequently dried and autoradiographed. Cold competitor assays were carried out adding a 100-fold molar excess of homologous unlabeled oligonucleotides of the various labeled doublestranded DNA probes. Supershift assays were performed by preincubation of nuclear extracts with 2 g of polyclonal Abs against p65 (Rel A), p50, Fos-B, c-Fos, JunB, or c-Jun, obtained from Santa Cruz Biotechnology, in the presence of all components of the binding reaction described above for 1 h at 4°C.

MSU Crystals Lead to Chemokine mRNA Expression in B10R Murine
Macrophages-To evaluate the capacity of MSU crystals to induce chemokine mRNA expression, B10R murine macrophages were incubated for 2 h with increasing concentrations of MSU crystals (50 -500 g/ml), and changes of chemokine mRNA levels were monitored by RPA. As shown in Fig.  1A, cell treatment with MSU crystals resulted in a concentrationdependent up-regulation of four different chemokine transcripts, MIP-1␤, MIP-1␣, MIP-2, and MCP-1. When 100 g/ml MSU crystals were administered, mRNA increases were already detectable for MIP-1␣, MIP-1␤, and MCP-1 (ϳ3-fold in-crease over negative control) and were more significant for MIP-2 (ϳ7-fold increase). Maximal values for MIP-1␤ (ϳ9-fold), MIP-1␣ (ϳ5-fold), MIP-2 (ϳ22-fold), and MCP-1 (ϳ6-fold) were observed upon macrophage exposure to 500 g/ml MSU crystals (the highest concentration tested). In parallel, kinetic analyses were performed in order to establish the time required to obtain maximal chemokine modulation with an intermediate concentration of MSU crystals (100 g/ml). As depicted in Fig.  1B, chemokine mRNA accumulation occurred very rapidly (0.5 h posttreatment), reached its peak after 1-2 h, and transiently decreased thereafter up to 8 h. Subsequent RPA experiments were thus conducted by stimulating cells for 2 h with 100 g/ml MSU crystals.
MSU Crystal-inducible Activation of the ERK1/2 Pathway Is Required for Chemokine mRNA Up-regulation-Having observed that MSU crystals mediate increases in chemokine mRNA expression, we next attempted to identify the second messengers responsible for these effects. We initially examined the induction of ERK1/2 MAPK, which are known to play an important role in chemokine regulation in response to various proinflammatory stimuli (32,33), including MSU crystals (26). As illustrated in Fig. 2, MSU crystals led to a rapid and sustained phosphorylation of the immediate upstream activator of ERK1/2, MAPK kinase 1/2 (MEK1/2) ( Fig. 2A), and of ERK1/2 (Fig. 2B). Both MEK1/2 and ERK1/2 phosphorylation were already detectable after 15 min and remained detectable up to 4 h poststimulation. Next, to evaluate the involvement of the ERK1/2 signaling pathway in MSU crystal-dependent chemokine regulation, cells were incubated for 1 h with increasing concentrations of specific inhibitors directed either against MEK1/2 (PD 98059) or ERK1/2 (apigenin) before MSU crystal stimulation (2 h). It should be noted that the employed concen-trations of these compounds correspond to those previously reported to specifically block this signaling cascade. In this regard, Liu et al. (26), who evaluated the effects of MSU crystals in human monocytes, showed that 50 M PD98059 blocked ERK1/2 phosphorylation but had no effect on p38 MAPK phosphorylation. Similarly, we have previously demonstrated that PD 98059 (40 M) and apigenin (50 M) inhibited ERK1/2 phosphorylation but did not affect those of Jak2 Tyr 1007 /Tyr 1008 FIG. 1. MSU crystal-inducible macrophage chemokine mRNA expression. Macrophages were stimulated either with 50 -500 g/ml MSU crystals for 2 h (A) or with 100 g/ml MSU crystals over an 8-h period (B), and chemokine mRNA expression was monitored using a mCK5 multiprobe RPA system (left panels). Densitometric quantification of chemokine mRNA levels over negative control after normalization to GAPDH is shown (right panels). Solid bars, MSU crystals. Results are representative of one of three separate experiments.
FIG. 2. MSU crystals lead to MEK1/2 and ERK1/2 phosphorylation in murine macrophages. After cell lysis, total proteins from either untreated or MSU crystal-stimulated macrophages (0.25-4 h) were subjected to Western blotting. MEK1/2 (A) and ERK1/2 (B) phosphorylation status were revealed with phospho-MEK1/2 and phospho-ERK1/2 Abs, respectively. Equal protein levels were verified using anti-MEK1/2 and anti-ERK1/2 Abs. Results are representative of one of three independent experiments. and STAT1␣ Tyr 701 in interferon-␥-stimulated M (34). Based on these data and in agreement with other independent studies (35)(36)(37)(38), subtoxic and specific concentrations of inhibitors against the MEK1/2-ERK1/2 pathway were selected to perform the experiments presented below. As depicted in Fig. 3A, 20 M PD 98059 dramatically reduced mRNA levels of MIP-1␤ (ϳ90% inhibition) and MIP-2 (ϳ85%), and caused a partial but significant decrease of MIP-1␣ (ϳ21%) and MCP-1 (ϳ58%) transcripts. As expected, high concentrations (40 M) of the MEK1/2 inhibitor abolished the expression of all four chemokines. In line with these data, cell pretreatment with apigenin (40 M) abrogated the MSU-inducible mRNA expression of MIP-1␣, MIP-1␤, and MCP-1 and almost completely suppressed that of MIP-2 (ϳ95% decrease) (Fig. 3B). This set of experiments demonstrates that the ERK1/2 pathway is activated in response to MSU crystals and indicates that this signaling cascade is functionally relevant to the modulation of chemokine induction by MSU crystals in macrophages.
Involvement of NF-B in MSU Crystal-dependent Chemokine Induction-The NF-B complex is involved in the transcriptional activation of chemokine genes (32,39), including that of monocytic IL-8 in response to MSU crystals (26). Based on this evidence, we first set out to elucidate whether this nuclear factor was activated in B10R macrophages upon stimulation by MSU crystals. In order to be translocated into the nucleus, NF-B must be released of its cytoplasmic inhibitor inhibitory protein B (IB). This occurs following IB phosphorylation, ubiquitination, and ultimately proteolytic degradation (40). Therefore, we investigated the ability of MSU crystals to induce IB␣ phosphorylation, an indicator of NF-B activation. Results obtained by Western blot revealed a rapid (detectable after 15 min) and transient phosphorylation of IB␣ on its Ser 32 residue, which was accompanied by a decrease of IB␣ protein levels when maximal IB␣ phosphorylation was detected (1-2 h) (Fig. 4A). Next, to confirm that NF-B activation was taking place in response to MSU crystals, EMSA experiments were undertaken. As shown in Fig. 4B, in cells stimulated with MSU crystals, NF-B nuclear translocation was observed at 30 min posttreatment, peaked after 1 h, and progressively declined thereafter up to 4 h. To define the nature of the MSU crystal-induced NF-B complex, supershift assays were performed using Abs directed against p50 and p65, two ubiquitous members of the NF-B family. As illustrated in Fig. 4C, the complex binding was diminished and partially supershifted in the presence of an anti-p50 Ab and almost completely abrogated by an anti-p65 Ab. Thus, MSU crystals appear to activate DNA binding of a p50/p50 homodimer (lower band) and a p50/p65 heterodimer (upper band) in murine macrophages. The specificity of these binding complexes was demonstrated by the fact that unlabeled NF-B oligonucleotide (100ϫ specificity) could compete effectively for binding, whereas an unrelated Sp1 probe (100ϫ nonspecific) could not. These data prompted us to define the functional importance of NF-B on MSU crystaldependent chemokine up-regulation. Knowing that selective blockage of the ERK1/2 pathway abolished chemokine expression by MSU crystals, we were interested to determine whether NF-B activation was also under the control of this signaling cascade. As depicted in Fig. 5A, cell exposure to either apigenin or PD 98059 resulted in a concentration-dependent decrease of MSU crystal-inducible NF-B nuclear translocation. To more directly address the putative contribution of NF-B to chemokine modulation, macrophages were incubated for 1 h with increasing concentrations (0.5-5 M) of BAY 11-7082, a chemical compound that blocks NF-B expression by inhibiting IB␣ phosphorylation (41), before MSU crystal stimulation, and NF-B translocation was monitored by EMSA (Fig. 5B). As expected, cells incubated with BAY 11-7082 showed a concentration-dependent down-regulation in the binding of the NF-B complex, which was nearly abolished in the presence of maximal inhibitor concentrations (5 M). Based on this control experiment, we next eval-

. MSU crystals induce IB␣ phosphorylation and NF-B nuclear translocation in murine macrophages.
A, following cell exposure to MSU crystals over a 4-h period, protein lysates were subjected to Western blotting to evaluate IB␣ phosphorylation status (upper panel). Changes in IB␣ protein levels were monitored using an anti-IB␣ Ab (lower panel). B, nuclear extracts from macrophages either left untreated or stimulated with MSU crystals for different time periods (0 -4 h) were incubated with a ␥-32 P-labeled NF-B probe and were subjected to EMSA. C, for supershift assays, nuclear proteins from cells stimulated with MSU crystals (100 g/ml for 1 h) were incubated or not with specific Abs against the p50 and p65 NF-B isoforms for 1 h at 4°C before EMSA. Binding specificity was tested by adding to nuclear extracts from 1-h treated cells a 100-fold molar excess of either a cold NF-B consensus oligonucleotide (100ϫ specific) or a nonspecific Sp1 probe (100ϫ nonspecific). These results are representative of one of three independent experiments. all chemokine transcripts, following the same pattern of inhibition as that exerted on NF-B (Fig. 5B). In fact, intermediate concentrations of this compound (3 M), which substantially diminished NF-B binding activity, caused a significant reduction of chemokine mRNA (ϳ65% for MIP-1␤, ϳ36% for MIP-1␣, ϳ53% for MIP-2, and ϳ64% for MCP-1). In line with these observations, cell exposure to 5 M BAY 11-7082, which suppressed NF-B translocation, also led to a total inhibition of the MIP-1␣, MIP-1␤, and MCP-1 transcripts and almost completely abrogated MIP-2 mRNA expression (ϳ94% reduction). Altogether, these data indicate that ERK1/2-mediated NF-B activation appears to be necessary for the induction of macrophage chemokine mRNA in response to MSU crystals.
MSU Crystals Induce AP-1 Nuclear Translocation via the ERK1/2 Pathway-In addition to evaluating the potential role of NF-B, we investigated the implication of AP-1 in MSU crystal-mediated chemokine modulation. This transcription factor is activated by MSU crystals in human monocytes and is involved in IL-8 regulation in response to this proinflammatory agent (26). Thus, we initially examined whether MSU crystals led to AP-1 activation in B10R macrophages. As depicted in Fig. 6A, when oligonucleotides containing AP-1 consensus binding sequences were used to probe nuclear extracts from MSU crystal-stimulated cells, we observed a rapid (30 min poststimulation) and sustained binding activity of this transcription factor, reaching its maximal expression at 4 h. To identify the AP-1 subunits that form the MSU crystal-inducible complex, supershift assays were conducted by incubating nuclear extracts from MSU crystal-treated macrophages with specific Abs against some of the main members of the AP-1 family: Fos B, c-Fos, JunB, and c-Jun. As illustrated in Fig. 6B, the AP-1 complex binding was diminished in the presence of Abs against c-Fos, JunB, and c-Jun, but it was not affected by an anti-FosB Ab. In parallel, the specificity of this binding complex was demonstrated by the fact that unlabeled AP-1 oligonucleotide (100ϫ specific) could compete effectively for binding, whereas an unrelated Sp1 probe (100ϫ nonspecific) could not. These data indicate that in macrophages, MSU crystals lead to nuclear translocation and binding activity of heterodimeric AP-1 complexes composed of c-Fos, JunB, and c-Jun. The similar kinetics of stimulation of AP-1 translocation and chemokine mRNA expression suggested a possible role for this transcription factor in MSU crystal-mediated chemokine regulation. To address this question, we next evaluated the involvement of the ERK1/2 pathway on the noticed AP-1 binding activity. When cells were treated with either apigenin or PD 98059 (Fig. 6C), AP-1 nuclear translocation in response to MSU crystals was reduced in a concentration-dependent manner. Of interest, the AP-1 complex DNA binding capacity was abrogated at the same inhibitor concentrations that were found to block chemokine expression (Fig. 3). Altogether, these results suggest that MSU crystal-induced chemokine up-regulation in macrophages involves the participation of ERK1/2-mediated AP-1 transcription factor activation.
MSU Crystal-mediated Increase of Chemokine mRNA Levels Is Due to Both Transcriptional and Posttranscriptional Controls-Previous studies performed by others (39,42) and by us (32) have indicated that both transcriptional and posttranscriptional levels of control modulate chemokine mRNA expression. Therefore, we were interested in establishing which of these mechanisms were responsible for the augmentation of macrophage chemokine mRNA levels following MSU crystal stimulation. To do so, we tested the effects of the transcriptional inhibitor actinomycin D (AD), as we described elsewhere (32). Initially, to verify whether MSU crystal-dependent chemokine regulation occurred at the transcriptional level, B10R cells were stimulated with MSU crystals for 2 h either in the absence or in the presence of AD (5 g/ml), and chemokine mRNA levels were monitored by RPA. As shown in Fig. 7A, AD treatment completely blocked the expression of all four chemokine transcripts by MSU crystals, indicating that, at least in part, MSU crystals regulate chemokine expression at the transcriptional level. Next, the ability of MSU crystals to increase chemokine mRNA stability was monitored by measuring their effects on the half-life of the various chemokine transcripts. To test this, cells were preincubated or not with MSU crystals for 2 h, and then they were exposed to AD (5 g/ml) over a 4-h FIG. 6. MSU crystals lead to AP-1 nuclear translocation via the ERK1/2 pathway. A, labeled AP-1 probe was incubated with nuclear extracts from cells either untreated or incubated with MSU crystals for different time periods (0 -4 h), and EMSA analysis was performed. B, for supershift assays, nuclear proteins from cells stimulated with MSU crystals (100 g/ml for 2 h) were incubated or not with specific Abs against one of the various AP-1 subunits, FosB, c-Fos, JunB, or c-Jun, for 1 h before EMSA. Binding specificity was tested by adding to nuclear extracts from 2-h treated macrophages a 100-fold molar excess of either a cold AP-1 consensus oligonucleotide (100ϫ specific) or a nonspecific Sp1 probe (100ϫ nonspecific). C, nuclear proteins from MSU crystal-stimulated cells (100 g/ml for 2 h) pretreated or not with 5-40 M apigenin or 20 -40 M PD 98059 were subjected to EMSA following incubation with an AP-1 consensus probe. Binding specificity was tested by adding to nuclear extracts from 2-h treated cells a 100-fold molar excess of cold AP-1 oligonucleotide. These results are representative of one of three independent experiments.
period. As depicted in Fig. 7B, in the presence of AD, chemokine mRNA from nonstimulated cells (control) decayed rapidly, with a half-life of ϳ30 min. MSU crystal treatment stabilized the MIP-1␣, MIP-2, and MCP-1 mRNAs, increasing their halflives to more than 4 h. In contrast, the half-life of the MIP-1␤ transcript was not enhanced by MSU crystals. Overall, these results indicate that MSU crystal-inducible MIP-1␤ mRNA expression is controlled only at the transcriptional level, whereas the regulation of MIP-1␣, MIP-2, and MCP-1 seems to be due to both transcriptional activation of their genes and posttranscriptional stabilization of their mRNA transcripts. DISCUSSION Acute gouty inflammation is characterized by a massive influx of leukocytes (mostly neutrophils) into the inflamed joints of hyperuricemic patients (2), which is mediated by the production of powerful chemoattractants and activators (8, 10 -15). Notably, monocyte IL-8 production has been proposed as a major mechanism mediating MSU crystal-induced neutrophil migration (6,10,11). Further supporting a role for mononuclear phagocytes as a source of chemokines during gout, we provide evidence for the induction of multiple CC and CXC chemokine transcripts (MIP-1␣, MIP-1␤, MCP-1, and MIP-2) in macrophages following stimulation by MSU crystals. Investigation of the transductional mechanisms responsible for this cellular response suggests the requirement of ERK1/2-dependent signals leading to nuclear factor activation and subsequent chemokine mRNA expression. In addition, our data from tran-scription and RNA decay assays indicated that the observed increase of macrophage chemokine mRNA levels occurs at the transcriptional level for MIP-1␤, whereas for MIP-1␣, MIP-2, and MCP-1 both transcriptional and posttranscriptional events appear to be involved.
As described by Liu and colleagues (26), ERK1/2-dependent signals are required for MSU crystal-inducible IL-8 mRNA accumulation in human monocytes. Extending these previous findings and in agreement with several studies linking ERK1/2 activity to chemokine regulation (32,33,43), our data indicated that the ERK1/2 pathway seems to play an important role in MIP-1␣, MIP-1␤, MIP-2, and MCP-1 mRNA expression by MSU crystal-stimulated macrophages. Although activation of the p38 MAPK pathway was also detected, specific inhibition of this signaling cascade had no effect on chemokine mRNA induction (data not shown), suggesting that despite the activation of at least two different MAPK, only the ERK1/2 pathway participates in this regulatory event. As to the mechanisms involved, experimental evidence suggested that this process implicates ERK1/2-mediated activation of two transcription factors, NF-B and AP-1. Indeed, the role of these factors in the transcriptional control of chemokine genes has been extensively documented (26,33,39,44) and has been linked to ERK1/2 activation (26,32). These results are perfectly in line with our published data showing that the up-regulating effect of MSU crystals on interferon-␥-mediated nitric oxide production occurred via ERK1/2-dependent NF-B activation (29). Thus, it is likely that through the activation of the same second messengers, MSU crystals modulate different macrophage functions, including multiple chemokine expression and amplification of macrophage responses to interferon-␥ (e.g. nitric oxide production), which would in turn contribute to the pathology related to gouty arthritis. Although the kinases through which the ERK1/2 pathway enhances NF-B translocation need to be identified, given that MAPK/ERK kinase kinase 1 has the ability to induce IB kinase activation (45), and both ERK1/2 and IB kinase were required for MSU crystaldependent NF-B binding to the IL-8 promoter (27), it is conceivable that the ERK1/2 pathway participates in IB kinase phosphorylation and subsequent MSU crystal-inducible NF-B nuclear translocation and chemokine transcription. Further investigation will bring light on this important matter.
Our data showing that chemokine mRNA up-regulation in response to MSU crystals occurs both at the transcriptional and posttranscriptional levels are in agreement with previous reports by others (39,42) and by us (32), in which the same mechanisms were found to control chemokine expression. Different lines of evidence indicate that this phenomenon could be associated with the presence of AU-rich motifs implicated in chemokine mRNA destabilization (39, 46 -49), and at least two mechanisms could be involved. First, members of the MAPK family, including c-Jun N-terminal kinase (50) and ERK1/2 (51,52) have been found to enhance the stability of certain mRNAs that bear AU-rich motifs. Since, according to previous studies (26,27) and our present results, MAPK activation takes place in MSU crystal-treated cells, these signaling cascades are likely to participate in the observed increases of chemokine mRNA half-lives. Second, it was postulated that enhancement on mRNA stability could be caused by the binding of a redoxsensitive protein to AU-rich motifs, which in turn form stable complexes that prevent mRNA degradation (42,53). Because MSU crystals are able to induce oxidative stress in monocytes (54), the possibility that this protein is activated should not be ruled out.
We found that MSU crystals rapidly and transiently upregulate mRNA expression of MIP-2, a potent neutrophil chemoattractant and activator (55). These observations are perfectly in line with in vivo studies reporting early neutrophil accumulation and increased MIP-2 production in a murine air pouch model of MSU crystal-mediated inflammation (8,15). Thus, based on these previous works, along with our data, it is plausible that MSU crystal-inducible macrophage MIP-2 contributes to neutrophil infiltration during acute gout. Although neutrophils account for most of the leukocytes recruited in response to MSU crystals (ϳ90%), a much smaller but still significant monocyte population (ϳ10%) is also observed (8,12,29). The mechanisms explaining monocyte migration into synovial tissue are not clearly defined; however, accumulating evidence indicates that one potential candidate is MCP-1 (8,12), a strong monocyte chemoattractant and activator (4). Indeed, intra-articular injection of MSU crystals resulted in MCP-1 secretion, and its inactivation reduced MSU crystalinduced monocyte infiltration by ϳ45% (12). Similarly, leukocyte migration was accompanied by elevated concentrations of MCP-1 and, to a lesser extent, of MIP-1␣ in the air pouch of MSU crystal-treated mice (8). In agreement with these findings, we detected a concentration-and time-dependent increase of MCP-1, MIP-1␣, and MIP-1␤ mRNA levels in macrophages stimulated with MSU crystals. Therefore, it is conceivable that CC chemokines and, more specifically, MCP-1 are important monocyte recruiters during gouty arthritis. Because in asymptomatic gout the cellular infiltrate is predominantly mononuclear (56), by favoring monocyte infiltration into the joint, MSU crystal-induced macrophage CC chemokines could contribute to the spontaneous resolution of acute gout.
In summary, our study demonstrates that MSU crystals regulate simultaneous activation of various chemokine genes in macrophages through a mechanism that seems to require ERK1/2-dependent cellular signals leading to transcription factor activation and subsequent chemokine mRNA expression. Moreover, our data indicated that MSU crystals control chemokine induction at both transcriptional and posttranscriptional levels. Importantly, our findings suggest that by increasing CXC chemokine production, macrophages present in the joint might play a role in early neutrophil infiltration during acute gout. In parallel, the possibility that macrophages participate in the spontaneous resolution of an acute gout attack, by favoring monocyte infiltration into the synovial environment via CC chemokine up-regulation, deserves further investigation. Overall, the current study will contribute to understanding the molecular mechanisms underlying macrophage modulation by MSU crystals, which is of paramount importance to better define the role of this cellular type in gouty arthritis and to design new therapeutic strategies for the management of this disease.