The Role of Mitogen-activated Protein Kinase Phosphatase-1 in the Response of Alveolar Macrophages to Lipopolysaccharide ATTENUATION OF PROINFLAMMATORY CYTOKINE BIOSYNTHESIS VIA FEEDBACK CONTROL OF p38*

Mitogen-activated protein (MAP) kinases are critical mediators of innate immune responses. In response to lipopolysaccharide (LPS), MAP kinases are rapidly activated and play an important role in the production of proinflammatory cytokines. Although a number of MAP kinase phosphatases (MKPs) have been identified, their roles in the control of cytokine production have not been well defined. In the present report, we investigated the role of MKP-1 in alveolar macrophages stimulated with LPS. We found that LPS triggered transient activation of three MAP kinase subfamilies, ERK, JNK, and p38, in both immortalized and primary murine alveolar macrophages. MKP-1 was rapidly induced by LPS, and its induction correlated with the dephosphorylation of these MAP kinases. Blocking MKP-1 with triptolide prolonged the activities of both JNK and p38 in immortalized alveolar macrophages. Stimulation of primary alveolar macrophages isolated from MKP-1 -deficient mice with LPS resulted in a prolonged p38 phosphorylation compared with wild type alveolar macrophages. Accord-ingly, these MKP-1 -deficient alveolar

Bacterial pneumonia represents a serious challenge to the public health. In the United States alone, ϳ4 to 5 million cases of community-acquired pneumonia occur each year, accounting for 10 million physician visits, a half million hospitalizations, and ϳ45,000 deaths (1). Among cases of the community-acquired pneumonia, ϳ10% are caused by Gram-negative bacteria. Nosocomial pneumonia, which is primarily caused by Gram-negative bacteria and has a mortality rate of up to 30%, accounts for about 15% of all hospital-acquired infections (2). Thus, elucidation of the pulmonary immune responses to Gram-negative bacteria is crucial for the development of therapeutic strategies to prevent and alleviate lung injury.
Alveolar macrophages constitute the first line of immune defense against microbial agents that infiltrate the gas-exchanging airways (3,4). These cells serve important phagocytic, microbicidal, and secretory functions and play a prominent role in lung immunity by initiating inflammatory and immune responses. Alveolar macrophages are essential for the daily clearance of air-borne microbial infiltration and maintaining the sterility of the delicate alveolar surfaces for effective gas exchange. When invading pathogens overwhelm the innate host defenses and establish an infection in the lung, as in bacterial pneumonia, alveolar macrophages are capable of initiating profound inflammatory responses. As a critical part of the innate immune defense, alveolar macrophages produce an array of inflammatory mediators that orchestrate the recruitment of polymorphonuclear leukocytes from the pulmonary vasculature into the alveolar spaces for the effective eradication of the offending pathogens (5). Among the crucial inflammatory mediators produced by alveolar macrophages are proinflammatory cytokines, including tumor necrosis factor (TNF) 1 -␣, interleukin 1␤, and interleukin 6. Although these inflammatory cytokines are beneficial to pulmonary host defense, excessive production of these cytokines is involved in the pathogenesis of inflammatory lung injury, which can result in failure of lung function in severe cases (6).
Macrophage activation and the subsequent production of proinflammatory cytokines as responses to Gram-negative bacteria have been studied extensively (7). Lipopolysaccharide (LPS), a cell wall component unique to Gram-negative bacteria, forms complexes with the LPS-binding protein. These complexes then interact with CD14 and toll-like receptor-4, leading to the activation of a multitude of signaling cascades that ultimately result in the biosynthesis of a group of proinflam-matory cytokines (7,8). MAP kinases, including extracellular signal-regulated kinases (ERKs), c-Jun N-terminal kinases (JNKs), and p38, play crucial roles in this process. Carter et al. (9) have demonstrated that, in LPS-stimulated human alveolar macrophages, both p38 and ERK are required for the maximal production of TNF-␣ and interleukin-6. It has been reported that JNK is potently activated in response to LPS stimulation in primary rat alveolar macrophages (10), although the role of JNK in the macrophage responses to LPS has not been defined. Also poorly understood are the negative regulation of MAP kinases in alveolar macrophages and the mechanisms responsible for the termination of the biosynthesis of proinflammatory cytokines. In the present report, we examined the function and regulation of MAP kinase phosphatase (MKP)-1 in LPS-stimulated primary and immortalized alveolar macrophages. We found that MKP-1 is highly induced by LPS in alveolar macrophages, and its induction is associated with the inactivation of MAP kinases. Using primary alveolar macrophages isolated from MKP-1-deficient mice, we demonstrated that MKP-1 acts in vivo to attenuate the p38 MAP kinase cascade, thus inhibiting the biosynthesis of proinflammatory cytokines. We also found that MKP-1 is highly induced by a group of corticosteroids commonly prescribed for the treatment of inflammatory lung diseases. Moreover, the abilities of these corticosteroids to induce MKP-1 are correlated with their relative antiinflammatory potencies. Our studies suggested that MKP-1 plays an important role in the pulmonary innate immune defense and could be a novel target for the development of new anti-inflammatory drugs.

EXPERIMENTAL PROCEDURES
Animals-Male C57BL/6 mice (8 -12 weeks old) were purchased from Harlan Sprague-Dawley (Indianapolis, IN). The generation of MKP-1 knock-out mice has been described previously (11). Cryopreserved embryos of MKP-1 knock-out mice (MKP-1 ϩ/Ϫ and MKP-1 Ϫ/Ϫ ) were kindly provided by Bristol-Myers Squibb Co. and were regenerated into mice in The Jackson Laboratory (Bar Harbor, ME). These mice were bred in-house to yield both wild type and MKP-1 Ϫ/Ϫ mice. All of the mice were maintained on Harlan Tecklad irradiated diet (Harlan) at 24°C with relative humidity between 30 and 70% on a 12-h day-night cycle. All of the animals received humane care in accordance with the guidelines of the National Institutes of Health and were sacrificed by CO 2 inhalation. The experimental protocols were approved by the Institutional Animal Care and Use Committee of the Columbus Children's Research Institute.
Isolation of Primary Murine Alveolar and Peritoneal Macrophages, Cell Culture, and Treatment-Resident bronchoalveolar macrophages were isolated from naive male C57BL/6, MKP-1 ϩ/ϩ , or MKP-1 Ϫ/Ϫ mice by alveolar lavage. Briefly, murine lungs were filled and flushed 15-20 times with 0.7 ml of prewarmed phosphate-buffered saline supplemented with 5 mM EDTA. This procedure yielded about 10 ml of lung lavage per animal. Cells in the alveolar lavage were collected by centrifugation at 800 ϫ g for 8 min at 4°C and plated into tissue culture dishes in RPMI 1640 (Mediatech, Herndon, VA) supplemented with 10% fetal bovine serum (Hyclone Laboratories, Logan, UT) and 2 mM L-glutamine. For immunofluorescence studies, the alveolar macrophages were seeded onto poly-D-lysine-coated coverslips. For assaying the role of MKP-1 in TNF-␣ production, cell suspensions containing 3 ϫ 10 5 alveolar macrophages, together with residual nonadherent cells, were plated into 6-well tissue culture plates. Four h later, nonadherent cells were removed by extensive washing with phosphate-buffered saline. The macrophages were cultured in complete medium overnight before being stimulated with 100 ng/ml LPS (Escherichia coli 055:B55, Calbiochem). After stimulation, cells were harvested into lysis buffer and culture medium collected as described previously (12).
Resident peritoneal macrophages were isolated from naive male MKP-1 ϩ/ϩ or MKP-1 Ϫ/Ϫ mice by carrying out a lavage of the peritoneum three times each with 4 ml of complete medium. Cells in the peritoneal lavage were collected by centrifugation at 800 ϫ g for 8 min at 4°C and were plated in RPMI 1640 supplemented with 5% fetal bovine serum and 2 mM L-glutamine. Four h after plating, nonadherent cells were removed by extensive washing with phosphate-buffered saline. The macrophages were cultured overnight in complete medium supple-mented with interferon-␥ (50 units/ml, CalBiochem) before being stimulated with LPS (100 ng/ml).
Mouse alveolar macrophage MH-S cells were purchased from ATCC (Manassas, VA) and were grown in complete medium (RPMI 1640, 10% fetal bovine serum, 2 mM L-glutamine, 1.5 g/liter NaHCO 3 , 4.5 g/liter glucose, 10 mM HEPES, 1 mM sodium pyruvate, and 50 M mercaptoethanol) at 37°C in a humidified atmosphere containing 5% CO 2 . LPS was added to the cell culture to achieve a final concentration of 100 ng/ml. Triptolide (Calbiochem), dissolved in Me 2 SO, was added to the medium at indicated concentrations 30 min before stimulating the cells with LPS. Dexamethasone, betamethasone, budesonide, flunisolide, triamcinolone, beclomethasone, clobetasol propionate, and prednisone, all purchased from Sigma, were dissolved in Me 2 SO as 5 mM stock solutions and were added to the medium to achieve a final concentration of 5 M.
Advenovirus Infection-The replication-deficient adenovirus expressing MKP-1 has been described previously (13). Null adenovirus was kindly provided by Dr. Michael Crow (Johns Hopkins University School of Medicine). On the day of viral infection, medium was removed, and cells, grown in 35-mm dishes, were washed once with medium containing 2% fetal bovine serum. The cells were incubated with adenovirus in 300 l of medium containing 2% fetal bovine serum for 1 h on a Belly Dancer (Stovall Life Science) shaking at 37°C in a humidified atmosphere containing 5% CO 2 , before complete medium was added to the culture. Twenty-four h later, cells were fed with fresh complete medium and were allowed to grow overnight before stimulation with LPS.
Western Blotting, ELISA, and Immunofluorescence-Western blot analysis was performed as described previously (12). MKP-1 was detected using a rabbit polyclonal antibody purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Phosphorylated JNK, p38, and ERK were detected using rabbit polyclonal antibodies purchased from Cell Signaling Technology (Beverly, MA). Total p38 or ␤-actin levels in the cell lysates were detected using an antibody against total p38 (Santa Cruz Biotechnology) or ␤-actin (Sigma), respectively. TNF-␣ in the culture medium was determined using the mouse ELISA Ready-SET-Go kits (eBioscience, San Diego, CA) according to the manufacturer's recommendation.
To detect the phosphorylation levels of p38 in alveolar macrophages using immunofluorescence, cells were first fixed in freshly prepared 4% paraformaldehyde for 15 min at 4°C and then were blocked with Tris-buffered saline (TBS) containing 10% normal goat serum and 1% bovine serum albumin at room temperature for 1 h. After blocking, the cells were incubated at 4°C overnight with a rabbit monoclonal antiphospho-p38 or anti-phospho-ERK antibody at a concentration of 20 g/ml. The cells were washed twice with TBS containing 0.1% Triton X-100 and once with TBS for 15 min, then were incubated with an Alexa 488-conjugated goat anti-rabbit IgG secondary antibody (Molecular Probes, Eugene, OR) for 1 h at room temperature. After washing five times with TBS containing 0.1% Triton X-100, the cells were stained with 4,6-diamidino-2-phenylindole (DAPI) to visualize the nuclei and were examined under a Zeiss Axioskop microscope (Carl Zeiss, Inc., New York). The fluorescent images were acquired with identical exposure time for all of the samples. The green fluorescent intensity was measured using AxioVision 3.1 software (Carl Zeiss). The fluorescent intensities of individual cells were defined as those above the background. Mean and S.E. were calculated from the fluorescent intensities of at least 20 randomly chosen cells.
Northern and Southern Blotting-Total RNA was isolated using STAT-60 (Tel-Test, Friendswood, TX). Northern blot analysis was carried out using a full-length mouse MKP-1 cDNA as a probe as described previously (12,14). The membrane was stripped and reprobed with 18 S rRNA to normalize for RNA loading.
For Southern blot analysis, genomic DNA was isolated from mouse tails according to standard procedures (15). Ten g of the genomic DNA was digested overnight with BamHI restriction enzyme. Southern blot analysis was performed using the full-length MKP-1 cDNA as a probe, as described previously (16).
Immune Complex Kinase Assays-JNK1 activity was measured by immune complex kinase assays as described previously (17)(18)(19), using [␥-32 P]ATP and recombinant GST-c-Jun (1-143) as a substrate (20). The incorporation of 32 P into the substrates was quantitated using a Storm system and the ImageQuant software (Amersham Biosciences).

Association between MKP-1 Induction and Inactivation of MAP Kinases in LPS-stimulated Alveolar Macrophages-To
examine the regulation of MAP kinases in alveolar macrophages, we stimulated immortalized murine alveolar macrophage MH-S cells with LPS for different periods of time. The activation kinetics of ERK and p38 MAP kinases in the cells were assessed by Western blot analysis using phospho-specific antibodies (Fig. 1A). In control alveolar macrophages, p38 activity was virtually undetectable. Upon LPS stimulation, p38 was rapidly activated, reaching maximal levels at about 15 min; then p38 activity promptly plummeted, returning to nearly basal levels within 45 min. The change in p38 was due to reversible phosphorylation, because total p38 levels did not change over the time course of the experiment. Likewise, ERK was also potently activated in response to LPS stimulation with kinetics similar to that of p38 activation. The effect of LPS on JNK1 activity was examined by immune complex kinase assays using GST-c-Jun as a substrate (Fig. 1B). Similar to what was observed for p38, JNK1 activity was also transiently activated in response to LPS, reaching maximal levels at 15 min and returning to base line after 2 h (Fig. 1B, lower panel). The lack of change in total JNK1 protein content indicates that JNK1 activation was also due to reversible phosphorylation (Fig. 1B, upper panel).
It has been shown that in mammalian cells, inactivation of MAP kinases is primarily carried out by a group of dualspecificity MAP kinase phosphatases (21). We have recently found that LPS induces the expression of MKP-1 in RAW264.7 cells (12). To examine whether MKP-1 plays a role in the regulation of MAP kinases in alveolar macrophages, we investigated the effect of LPS on MKP-1 expression. MKP-1 mRNA was examined with Northern blot analysis ( Fig. 2A). Unstimulated MH-S alveolar macrophages expressed low basal levels of MKP-1 mRNA ( Fig. 2A). Upon LPS stimulation, MKP-1 mRNA levels were rapidly elevated. The increase in MKP-1 mRNA expression was clearly evident within 15 min. Maximal MKP-1 mRNA levels were seen at about 60 min, followed by a gradual decline. Mirroring the increases in MKP-1 mRNA levels, MKP-1 protein levels were also increased upon LPS stimulation (Fig. 2B). MKP-1 protein band was clearly visible within 30 min, and its levels were significantly elevated by 45 min. Maximal MKP-1 protein levels were reached at about 60 min. Although MKP-1 mRNA levels decreased to nearly basal levels by 120 min (Fig. 2A), MKP-1 protein levels subsided only moderately (Fig. 2B). Comparable sample loading was confirmed using an antibody against a housekeeping protein, ␤-actin (Fig.  2B). The tight correlation between MKP-1 induction and the inactivation of MAP kinases suggests that MKP-1 plays a significant role in mediating the dephosphorylation of these MAP kinases.
To further explore the relationship between MKP-1 and MAP kinases, we used triptolide to pharmacologically suppress the induction of MKP-1 and examined the effect of this MKP-1 blockade on the activation kinetics of MAP kinases after LPS stimulation. Pretreatment of MH-S cells with triptolide did not exhibit an appreciable effect on the basal MKP-1 levels. Triptolide attenuated LPS-induced MKP-1 accumulation in a dosedependent manner (Fig. 3A, upper panel). Consistent with the

FIG. 1. LPS induces a transient activation of MAP kinases in MH-S alveolar macrophages.
MH-S alveolar macrophages were stimulated with 100 ng/ml LPS. A, activation kinetics of ERK and p38 after LPS stimulation. The phosphorylation of ERK and p38 MAP kinases was examined by Western blot analysis using antibodies against phospho-p38 (p-p38) and phospho-ERK (p-p44 ERK1 , p-p42 ERK2 ). Antibody against p38 (p38) was used to verify comparable sample loading. B, time course of JNK activation. JNK1 activity was determined by immune complex kinase assays using [␥-32 P]ATP and recombinant GST-c-Jun as a substrate (lower panel). The activity was quantitated using a PhosphorImager and the ImageQuant program and expressed as -fold increase relative to untreated control. Total JNK1 protein was detected by Western blot analysis, serving as a control for sample input (upper panel). idea that MKP-1 plays an important role in the dephosphorylation of stress-activated MAP kinases, the blockade of MKP-1 induction by triptolide considerably delayed the dephosphorylation of p38 and JNK (Fig. 3, A and B). In contrast, triptolide exhibited little effect on the dephosphorylation of ERK (Fig.  3A). Triptolide pretreatment neither prevented nor delayed the responses of alveolar macrophages to LPS, as indicated by the prompt activation of p38 in the presence of triptolide (Fig. 3C). These results indicate that triptolide did not postpone the initiation of LPS-triggered signal transduction, arguing that MKP-1 blockade was responsible for the prolonged activation of p38 and JNK in LPS-stimulated MH-S cells.
To investigate whether MKP-1 plays a significant role in the responses of primary alveolar macrophages to LPS, we studied MAP kinase regulation using bronchoalveolar macrophages isolated from naive male C57BL/6 mice. We found that recovery of alveolar macrophages from the lungs by lavage was relatively small; one adult mouse yielded ϳ3-5 ϫ 10 5 alveolar macrophages. To overcome the difficulty in obtaining large quantities of primary alveolar macrophages, we performed immunofluorescence assays to assess the activation of p38 in LPS-stimulated primary alveolar macrophages, using a monoclonal rabbit antibody specific for phosphorylated p38 (Fig. 4A). In unstimulated alveolar macrophages, low basal levels of phospho-p38 were detected. Upon LPS stimulation, phospho-p38 levels were markedly elevated within 15 min, with maximal activity attained at about 30 min, which was followed by a gradual decline in phospho-p38 levels. It is worth noting that at 15 min the phospho-38 levels in the cells appeared to vary widely, probably reflecting the heterogeneity in the responses of individual cells. Quantitation of the fluorescent intensity in individual cells, as a measurement of phospho-p38 levels, indicated that p38 activity decreased by ϳ58% after 1 h and dropped by ϳ70% after 2 h (Fig. 4A). Phosphorylated (that is, active) p38 appeared to localize in both the cytosolic and nuclear compartments, which were distinguished by DAPI staining (Fig. 4A). To validate the methodology of the immunofluorescence-based p38 activity assay, we performed Western blot analysis to examine the activation kinetics of p38 after LPS stimulation (Fig. 4B). From 20 C57BL/6 mice, ϳ8 ϫ 10 6 alveolar macrophages were collected and pooled. Similar to what was observed with the immunofluorescence assays, LPS stimulation resulted in a transient activation of p38, with maximal phosphorylation levels observed between 15 and 30 min. After the peak, p38 activity rapidly descended by 60 min. Likewise, JNK and ERK also underwent a transient activation with kinetics similar to those of p38 activation (Fig. 4B). Correlated with the inactivation of these MAP kinases, MKP-1 underwent a potent induction upon LPS stimulation (Fig. 4B). MKP-1 protein became clearly evident by 30 min and reached high levels by 60 min. Immunofluorescent analysis of phospho-ERK in fixed cells confirmed the observations made by Western blot analysis (Fig. 4C). Taken together, our results validated the methodology of immunofluorescence for the assessment of phospho-p38 levels in LPS-stimulated alveolar macrophages. More importantly, these results strongly suggest that MKP-1 is a physiologically relevant regulator of MAP kinases in alveolar macrophages during bacterial infection.
Effects of MKP-1 Deficiency on p38 Activity and TNF-␣ Biosynthesis in Alveolar Macrophages-To further elucidate the physiological functions of MKP-1, we obtained cryopreserved embryos of MKP-1 knock-out mice from Bristol-Myers Squibb and regenerated these embryos into mice. The MKP-1 transgenic mice were bred to produce wild type (MKP-1 ϩ/ϩ ), heterozygous (MKP-1 ϩ/Ϫ ), and homozygous (MKP-1 Ϫ/Ϫ ) offspring. The genotypes of these progenies were determined by Southern blot analysis (Fig. 5A). To verify the deficiency of MKP-1 protein in the MKP-1 knock-out mice, resident peritoneal macrophages were isolated from wild type and MKP-1 Ϫ/Ϫ mice and primed with interferon-␥ overnight. Stimulation of the peritoneal macrophages with LPS for 60 min resulted in a large increase in MKP-1 protein in wild type but not in MKP-1 Ϫ/Ϫ cells, indicating the lack of MKP-1 function in these animals (Fig. 5B). Importantly, compared with the MKP-1 ϩ/ϩ peritoneal macrophages, the MKP-1 Ϫ/Ϫ peritoneal macrophages exhibited significantly elevated levels of phospho-p38, phospho-JNK, and phospho-ERK proteins, illustrating a crucial role for MKP-1 in the dephosphorylation of these kinases in these cells (Fig. 5B). Because of the relatively limited numbers of alveolar macrophages isolated from mice, the kinetics of p38 activation in wild type and MKP-1-deficient alveolar macrophages were examined by immunofluorescence using an antibody specific for phospho-p38 (Fig. 5C). In wild type alveolar macrophages, LPS resulted in a transient p38 phosphorylation that reached its maximal levels at 15-30 min and returned to nearly basal levels by 1 h. LPS also triggered a rapid p38 phosphorylation in MKP-1-deficient cells. However, unlike the rapid dephospho-

FIG. 3. Blocking MKP-1 induction by triptolide prolongs LPSelicited MAP kinase activation in MH-S alveolar macrophages.
MH-S cells were pretreated with triptolide at the indicated concentrations for 30 min before the addition of LPS (100 ng/ml). A, dose-response effect of triptolide on MKP-1 expression and on the phosphorylation of ERK (p-ERK) and p38 (p-p38). Cells that received no triptolide were pretreated with equal volumes of Me 2 SO. Total p38 (p38) served as a loading control. B, dose-response effect of triptolide on JNK activity. JNK1 activity was determined by immune complex kinase assays using [␥-32 P]ATP and recombinant GST-c-Jun as a substrate. The graph depicts -fold changes in JNK1 activity relative to untreated control. C, effect of triptolide-mediated MKP-1 blockade on the kinetics of p38 phosphorylation (p-p38). Western blot of total p38 (p38) served as a control for sample loading. rylation of p38 observed in wild type alveolar macrophages, p38 activity persisted for a substantially longer time in MKP-1deficient alveolar macrophages. In fact, the levels of p38 phosphorylation at 60 and 120 min did not markedly differ from the peak levels (Fig. 5C). These results indicate that MKP-1 plays a critical role in the negative feedback control of p38 in LPSstimulated alveolar macrophages.
To understand the function of MKP-1 in the regulation of TNF-␣ production, alveolar macrophages isolated from both wild type and the MKP-1-deficient mice were stimulated with LPS for 4 and 6 h, and the medium was harvested. TNF-␣ concentrations in the medium were measured by ELISA. Compared with the wild type alveolar macrophages, the MKP-1deficient cells produced significantly more TNF-␣ at both the 4and 6-h time points (Fig. 6A). Our results indicate that MKP-1-deficient alveolar macrophages were able to mount a more robust and rapid inflammatory responses than were wild type cells.
To further define the function of MKP-1 in the regulation of proinflammatory cytokines in alveolar macrophages, we examined the effect of MKP-1 overexpression on TNF-␣ production in MH-S alveolar macrophages. Because MH-S cells were extremely difficult to transfect, adenovirus was utilized to mediate high level MKP-1 expression. MH-S cells were infected with either null or MKP-1-expressing adenoviruses at doses of 200, 400, or 800 multiplicities of infection (m.o.i.). Thirty-six h later, cells were stimulated with 100 ng/ml LPS for 4 or 6 h. TNF-␣ secreted into the medium was determined by ELISA (Fig. 6B). Infection of cells with the MKP-1-expressing adenovirus, but not the null adenovirus, resulted in a dose-dependent reduction of TNF-␣ secretion. Infection with 800 m.o.i. of MKP-1 adenovirus resulted in a 53% reduction in TNF-␣ production, whereas infection with the same m.o.i. of null virus had no significant effect. These results indicate that enhanced expression of MKP-1 inhibits TNF-␣ production. Taken together, these results suggest that MKP-1 plays an important role in modulating the responses of alveolar macrophages to LPS to prevent overproduction of proinflammatory cytokines.
Induction of MKP-1 by Anti-inflammatory Corticosteroids-Corticosteroids are potent anti-inflammatory drugs frequently prescribed to treat various inflammatory lung diseases, including asthma (22), chronic obstructive pulmonary disease, and acute respiratory distress syndrome (23). Corticosteroids strongly inhibit the production of inflammatory cytokines triggered by LPS (24). Corticosteroids, such as dexamethasone, have been shown to block JNK, p38, and ERK activation in various cell types (25)(26)(27). Recent studies have shown that MKP-1 is induced by dexamethasone in a number of cell types, including macrophages, epithelial cells, and mast cells (12,26,28). To investigate the relationship between the clinical potencies of anti-inflammatory glucocorticoids and their abilities to induce MKP-1, we examined the induction of MKP-1 in MH-S alveolar macrophages by eight commonly prescribed anti-inflammatory corticosteroids: budesonide, flunisolide, triamcinolone, prednisone, beclomethasone, clobetasol propionate, dexamethasone, and betamethasone. MH-S cells were treated with each of the corticosteroids at a concentration of 5 M for 8 h, and MKP-1 mRNA was then examined by Northern blot analysis. All eight corticosteroids enhanced the expression of MKP-1, although to varying degrees (Fig. 7). Moreover, higher

FIG. 4. MKP-1 induction is tightly associated with dephosphorylation of MAP kinases in LPS-stimulated primary murine alveolar macrophages.
A, phosphorylation kinetics of p38. Primary alveolar macrophages isolated from male C57BL/6 mice were treated with LPS (100 ng/ml). p38 phosphorylation was analyzed by immunofluorescence using an antibody against phospho-p38. Green fluorescence represents phospho-p38 (upper panels). The cell nuclei were stained with DAPI to localize the cells (lower panels). All green fluorescent images were acquired under identical conditions: 1.5-s exposure, ϫ400 amplification. The intensity of green fluorescence was quantitated using AxioVision 3. flunisolide ϳ triamcinolone (30). Although direct comparisons of the topical steroid clobetasol propionate with other drugs are unavailable, clobetasol propionate is generally regarded by clinicians to be a highly potent anti-inflammatory corticosteroid. Thus, the results presented here strongly support the working hypothesis that MKP-1 may play a significant role in the antiinflammatory action of glucocorticoids. DISCUSSION MAP kinase pathways play an important role in orchestrating the immune responses to microbial infection (31). In macrophages, MAP kinases are crucial regulators of the production of proinflammatory cytokines in response to bacterial infection. Because the MAP kinases are inactivated through dephosphorylation mediated by a family of MKPs (21), it is conceivable that these phosphatases may play a significant role in the negative control of cytokine production. In this report, we have examined the role of an archetypical MKP, MKP-1, in the innate immune responses of alveolar macrophages to LPS, a model of Gram-negative bacterial infection. We found that in both immortalized and primary alveolar macrophages, MKP-1 was highly induced by LPS, and this induction correlated with dephosphorylation of all three MAP kinases, including ERK, JNK, and p38 (Figs. 1, 2, and 4). Triptolide completely blocked LPS-induced MKP-1 transcription (data not shown) and substantially attenuated the increase in MKP-1 protein. It also markedly slowed the time-dependent inactivation of JNK and p38 (Fig. 3). Adenovirus-mediated overexpression of MKP-1 in the immortalized MH-S alveolar macrophages significantly attenuated the production of TNF-␣, illustrating the capacity of MKP-1 to switch off the MAP kinase cascades and modulate the production of proinflammatory cytokines (Fig. 6). Most importantly, we found that the dephosphorylation of p38 that followed the initial activation after LPS treatment was significantly delayed in MKP-1-deficient alveolar macrophages compared with their wild type counterparts (Fig. 5). Moreover, MKP-1-deficient cells were able to mount a more robust and rapid production of TNF-␣ than were wild type cells in response to LPS (Fig. 6). These results indicate that MKP-1 may play a pivotal role in the negative feedback control of p38 in alveolar macrophages after LPS stimulation. Our studies provide strong evidence to support a physiological role for MKP-1 in restraining the production of inflammatory cytokines by the innate immune cells after pulmonary infection.
In transfection assays, MKP-1 is capable of dephosphorylating all three MAP kinase subfamilies, including ERK, JNK, and p38 (17,32,33). More recent studies have suggested that MKP-1 prefers p38 and JNK as substrates (34). This is consistent with a report that disruption of the MKP-1 gene does not significantly affect the activation kinetics of ERK in serumstimulated mouse embryonic fibroblasts (11). We found that in peritoneal macrophages, MKP-1 appears to regulate all three MAP kinase families, because MKP-1-deficient peritoneal macrophages exhibited higher phosphorylation levels for all three MAP kinases than did wild type cells after LPS stimulation (Fig. 6B). It still remains debatable as to whether MKP-1 plays a critical role in the dephosphorylation of all three MAP kinase subfamilies in alveolar macrophages. In MH-S alveolar macrophages, attenuation of LPS-induced MKP-1 accumulation by triptolide had no appreciable effect on ERK activity (Fig. 3), suggesting that ERK is probably not a target of MKP-1 in this cell line. Several limitations with the immortalized MH-S alveolar macrophages make them less than ideal to address this particular question. First, although triptolide nearly completely blocked the transcriptional induction of the MKP-1 gene, as indicated by Northern blot analysis (data not shown), triptolide did not completely prevent the increase in MKP-1 protein after LPS stimulation (Fig. 3), which is most likely due to the increase in MKP-1 protein stability (12,35). Second, MH-S alveolar macrophages are extremely difficult to transfect, restricting the genetic manipulation of these cells. Finally and perhaps most importantly, immortalization of alveolar macrophages might have altered the regulation of critical signal transduction pathways. To overcome these limitations, we took advantage of the existing MKP-1 knock-out transgenic mice and studied the physiological function of MKP-1 in primary alveolar macrophages. However, harvesting primary alveolar macrophages presented unique problems secondary to the relative scarcity of alveolar macrophages in mice and the small size of our MKP-1 knock-out mouse colony. Therefore, we employed immunofluorescence techniques to compare MAP kinase activities between primary alveolar macrophages obtained from wild type and those obtained from MKP-1-deficient mice. We found that immunofluorescent analysis of both phospho-p38 and phospho-ERK in a few dozen alveolar macrophages yielded results very similar to those obtained by Western blotting, indicating the reliability of such assays for these two kinases (Fig. 4). By using immunofluorescence, we provided unequivocal evidence supporting a critical role of MKP-1 in the negative feedback control of p38 activity in LPS-stimulated alveolar macrophages (Fig. 5C). However, immunofluorescent analysis of phospho-ERK levels so far has not conclusively revealed an increase in ERK activity in LPS-stimulated MKP-1-deficient alveolar macrophages (data not shown). To date, we have been unable to identify an antiphospho-JNK antibody that is suitable for immunofluorescence. We plan to clarify the role of MKP-1 in the regulation of JNK and ERK in LPS-stimulated alveolar macrophages in the future.
A potentially important and clinically relevant finding in this study is the correlation of the relative anti-inflammatory potencies of corticosteroids with their ability to induce MKP-1 mRNA expression. As effective anti-inflammatory drugs, corticosteroids are widely used to treat many inflammatory lung diseases. One such corticosteroid, dexamethasone, has been shown to inhibit the activities of MAP kinases (25,26,28) and to block the production of inflammatory cytokines (24). We have previously demonstrated that MKP-1 is potently induced by dexamethasone in RAW264.7 macrophages (12). In this report, we studied eight corticosteroids commonly prescribed for the treatment of inflammatory lung diseases and found a tight correlation between the relative anti-inflammatory potencies of these steroids and the ability to induce MKP-1 expression at a clinically relevant concentration. This is remarkable given that the anti-inflammatory potencies of the steroids are influenced by both the pharmacokinetic properties of the drugs and the molecular actions of the drugs within the cells. These results strongly suggest that induction of MKP-1 may contribute to the therapeutic activity of these drugs. We speculate that impairment of MKP-1 expression after bacterial infection may in part be responsible for the exaggerated inflammatory response seen in some lung diseases, such as asthma and acute respiratory distress syndrome. Therefore, MKP-1 may represent a novel therapeutic target for treating inflammatory lung diseases.