Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Peroxisome proliferator-activated receptors (PPARs) are nuclear hormone receptor transcription factors that regulate adipocyte differentiation and metabolism (1). Three PPAR isoforms (, , and ) have been identified to date. Ligands for PPAR- include naturally occurring compounds such as the hydroxyeicosatetraenoic acids (HETEs), hydroxyoctadecanoic acids (HODEs), and prostaglandin D₂ metabolite 15-deoxy-^12,14 prostaglandin J₂ (15d-PGJ₂) (2, 3), as well as synthetic compounds such as thiazolidinedione antidiabetic agents (4) and fibrates. Ligand-induced activation of PPAR results in heterodimerization of the receptor with the retinoid X receptor and subsequent binding to specific peroxisome proliferator-responsive elements located within the promoter region of target genes (5).

PPAR- has been shown to play a major role in regulating adipocyte differentiation and glucose homeostasis (6, 7). In addition, it has been proposed that thiazolidinediones may possess anti-inflammatory properties (8). In adipose tissue, the adipogenic action of PPAR- agonists are opposed by several proinflammatory cytokines, including tumor necrosis factor (TNF)- and interferon (IFN)-. In vitro, the antidiabetic thiazolidinediones block the effects of TNF- on both adipogenesis and insulin sensitivity. Recent studies demonstrate that 15d-PGJ₂ blocks IFN-- induced murine macrophage activation (9). These findings have prompted several laboratories to begin investigating the role of PPAR- as an immunomodulator.

Although the cellular expression pattern of PPAR- in pulmonary tissue has not been well characterized, several lines of evidence suggest that airway epithelium may also express PPAR- (10). In murine macrophages, expression of PPAR- is upregulated by interleukin (IL)-4, a cytokine believed to play a crucial role in certain subsets of airways inflammation (13). In the same study, IL-4 induced 12/15-lipoxygenase (12/15-LO), an enzyme capable of generating PPAR- agonists in vivo. 12/15-LOs are also highly expressed in surface airway epithelial cells under basal conditions (14).

Here, we demonstrate that airway epithelial cells constitutively express high levels of PPAR-. Activation of PPAR- dramatically inhibited the cytokine-induced expression of inflammatory mediators in airway epithelial cells, suggesting that PPAR- may act as a negative immunomodulator in the airways.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Recombinant IL-1, TNF-, and IFN- were obtained from R&D Systems (Minneapolis, MN), fetal calf serum (FCS) from GIBCO BRL (Gaithersburg, MD), and 15d-PGJ₂ and ciglitazone from Biomol (Plymouth Meeting, PA). (AOx)₃-TK-luciferase is a thymidine kinase/luciferase reporter gene containing three direct repeats of a PPAR response element cloned upstream of the thymidine kinase promoter and was a generous gift of Dr. Ron Evans (Salk Institute, La Jolla, CA). A -actin promoter/-galactosidase reporter gene was used for normalization of luciferase activity. The mouse monoclonal immunoglobulin G antihuman PPAR- antibody (E8/sc-7273) was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).

Cell Culture and Activation Protocol

NIH-A549, HeLa, BEAS-2B, and THP-1 cells were obtained from ATCC (Rockville, MD). A549 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 5% human serum (Gemini Bioproducts, Calabasas, CA). The other cell lines were grown in DMEM with 10% FCS with penicillin (100 U/ ml) and streptomycin (100 µg/ml). NIH-A549 cells were plated in assay medium (0.5% human serum) and then treated with or without PPAR- agonists and/or cytokine mix (10 ng/ml IL-1, 10 ng/ml TNF-, and 100 U/ml IFN-) for 24 h and assessed for inducible nitric oxide (NO) synthase (iNOS) expression and IL-8 secretion.

Immunoblot and Northern Analyses

NIH-A549 cells were harvested, washed, pelleted, and resuspended in lysis buffer. After protein concentrations were determined, Laemmli's loading buffer was added. Whole-cell extracts from A549 cells (50 µg/lane) were resolved on 10% sodium dodecyl sulfate-polyacrylamide gels and electroblotted to nitrocellulose. Membranes were then probed with anti-PPAR- E8 or iNOS antibodies (Transduction Laboratories, San Diego, CA) according to manufacturer's protocols.

PPAR-- and --specific probes were generated using Hifidelity Taq polymerase (Stratagene, Palo Alto, CA) according to the manufacturer's suggested protocols. Briefly, total RNA was prepared from NIH-A549 cell cultures using Trizol (Life Technologies, Gaithersburg, MD) and used in reverse transcription reactions. The resulting complementary DNA (cDNA) was then used as template in polymerase chain reaction (PCR) reactions using primers specific to PPAR- (upstream: atgaccatggttgacacaga; downstream: 5'gcagccctgaaagatgcgga3') and PPAR- (upstream: atggtggacacg gaaagccc; downstream: 5'gcagccctgaaagatgcgga3'). These primers were designed to amplify either PPAR-- or PPAR--specific sequence and not PPAR-. The amplified fragments were cloned into a sequencing vector and sequenced in both directions to confirm their identity to cloned cDNAs. Clones from two different reverse transcriptase-PCR reactions revealed a fragment identical to published human PPAR- sequence. Both clones contained the same GC- to -CG inversion at base pairs (bp) 727-728 (Figure 1). This inversion would cause nonconservative AA changes and is believed to be PCR artifacts.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Sequence comparison of the published human PPAR- cDNA nucleotide sequence with cDNA probe. Sequence alignment (nucleotides 719-733) demonstrating GC-to-CG inversion contained in the cDNA probe and the resulting amino acid change as compared with the published human PPAR- sequence (Accession #NM 005037).

Northern analysis was performed using standard techniques. Briefly, ³²P-labeled, human-specific PPAR- and PPAR- cDNA probes were prepared by random priming (specific activity approximately 10⁸ counts per min/ng). Total RNA was resolved on a 1.0% agarose/formaldehyde gel and transferred to nylon membrane overnight. Membranes were prehybridized for 20 min at 42°C and then hybridized with the labeled probes in QuikHyb (Stratagene). After hybridization, the membranes were washed as described in the manufacturer's protocols.

Transient Transfection Assays

NIH-A549 cells were transiently transfected using calcium phosphate according to standard techniques (15). Protein extracts generated from harvested cells were assessed for luciferase (Tropix, Bedford, MA) and -galactosidase (Promega, Madison, WI) activity using commercial chemiluminescence assays. Luciferase activity was normalized to -galactosidase activity.

IL-8 Enzyme-Linked Immunosorbent Assays

Corning-Costar (Acton, MA) 96-well plates were coated with a monoclonal goat antihuman IL-8 antibody (R&D Systems) overnight at 4°C and blocked with 0.1% bovine serum albumin. Cells were treated overnight with or without PPAR- and/or a mixture of inflammatory cytokines. Equal aliquots of conditioned media were then placed into wells. In some cases, the media was diluted up to 1:25 to insure that the concentration of IL-8 was within the range of the assay. After 2 h, a polyclonal rabbit antihuman IL-8 antibody was applied (Upstate Biotechnology, Saranac Lake, NY) followed by a biotinylated goat antirabbit secondary antibody (Vector, Burlingame, CA). Next, wells were incubated with streptavidin alkaline phosphatase (Jackson Immunoresearch, West Grove, PA) and detection of captured IL-8 complexes was achieved using an alkaline phosphatase substrate kit (Sigma, St. Louis, MO). Absorbance was measured at 405 nm using a Microplate Devices reader.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Airway Epithelial Cells Express PPAR- Protein and Messenger RNA

To determine whether human airway epithelial cells express PPAR-, we analyzed protein extracts from two airway epithelial cell lines, BEAS-2B and NIH-A549, by standard immunoblotting techniques using the E8/sc-7273 antibody. As a control, protein extracts from Hela cells transfected with an expression vector for murine PPAR- were analyzed concurrently. Using this approach, protein extracts from both BEAS-2B and NIH-A549 cells were found to contain a protein that comigrated with the PPAR- control (Figure 2A). BEAS-2B and NIH-A549 cells constitutively expressed high levels of PPAR- protein. NIH-A549 cells were selected for further studies because these cells have an intact signal transducer and activator of transcription (Stat) 6 signal transduction pathway. IL-4 treatment significantly upregulated PPAR- expression in NIH-A549 cells. Similar levels of expression were demonstrated in protein extracts of other airway cell lines, including 1HAE and 9HTE cells (data not shown).

View larger version (18K): [in this window] [in a new window]   Figure 2.   Expression of PPAR- in airway epithelial cells. (A) Protein extracts (50 µg/ lane) were prepared from BEAS-2B, THP-1, and NIH-A549 cells and analyzed using standard immunoblotting techniques and the E8/sc-7273 antibody at a 1:1,000 dilution. HeLa cells were transfected with an expression vector for murine PPAR- and used as a positive control. NIH-A549 cells were analyzed under basal conditions and after stimulation with IL-4 for 24 h. The blot was reprobed using a primary antibody to -actin as a control for protein loading. (B) Protein extracts were prepared and analyzed as in A. NIH-A549 cells were assessed at baseline and after 24 and 48 h of treatment with IL-4 (10 ng/ml) as indicated. (C) Total RNA (25 µg/lane) was isolated from NIH-A549 cells, IL-4 (10 ng/ml)-stimulated NIH-A549 cells, and THP-1 cells and analyzed using standard Northern analysis techniques. After immobilization, the RNA was probed with an 857-bp labeled DNA fragment that was specific for PPAR-. The filter was reprobed with a labeled full-length -actin cDNA probe to assess RNA loading.

Although the E8 antibody preferentially recognizes PPAR-, it cross-reacts with PPAR-. To confirm the identity of the PPAR- isoform, an 857-bp PPAR--specific probe (see MATERIALS AND METHODS) was used to perform Northern analysis using total RNA harvested from THP-1 and NIH-A549 cells. A single hybridization signal migrating at the appropriate size was detected (Figure 2B). The hybridization signal was much stronger in airway cells than in THP-1 cells. Further, IL-4 upregulated PPAR- messenger RNA (mRNA) expression by 2- to 3-fold in NIH-A549 cells. Parallel studies using a PPAR--specific probe did not identify a signal in mRNA from NIH-A549 or BEAS-2B cells (data not shown).

Endogenous Airway Epithelial PPAR- Is Functional

We next assessed whether the PPAR- expressed by NIH-A549 cells could be activated. NIH-A549 cells were transfected with a PPAR-dependent promoter gene construct, (AOx)₃-TK-luciferase. This luciferase reporter gene contains three direct repeats of the acyl COA (rat) PPAR response element subcloned upstream of the thymidine kinase promotor. It is well characterized, and the vector from which it was derived is not responsive to PPAR ligands (2, 16). After resting overnight in 0.5% human serum, the cells were treated with 1 or 5 µM 15d-PGJ₂ for 24 h, as indicated in the figures. These concentrations of 15d-PGJ₂ were selected because they are associated with the inhibition of inflammatory cytokine release in murine and human monocytes (17, 18). The data in Figure 3A demonstrate that 15d-PGJ₂ activates transcription of the (AOx)₃-TK-luciferase reporter gene in a dose-dependent manner.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Functional characterization of airway PPAR-. NIH-A549 cells were cotransfected with (AOx)3-TK-luciferase (2 µg) and a -actin promoter/-galactosidase (100 ng). After resting in 0.5% human serum overnight, the cells were stimulated for 24 h with 15d-PGJ2 (A) or ciglitazone (B) at the indicated concentrations. (C) NIH-A549 cells were cotransfected with (AOx)3-TK- luciferase (2 µg), a -actin promoter/-galactosidase (100 ng), and a murine PPAR- expression vector (100 ng) as indicated. After resting overnight in 0.5% human serum, the cells were stimulated with a PPAR- agonist, WY14643 (50 µM), for 24 h. (D) NIH-A549 cells were cotransfected with (AOx)3-TK-luciferase (2 µg) and a -actin promoter/-galactosidase (100 ng) expression vector. The cells were rested for 16 h in culture medium with 0.5% human serum containing IL-4 (10 ng/ml) and then stimulated with ciglitazone as indicated. Protein extracts were assayed for luciferase and -galactosidase activity as described. Luciferase activity was normalized to -galactosidase activity. The results are displayed as means of triplicates with a standard error. The results are representative of at least two experiments.

Several investigators have raised concerns about the use of 15d-PGJ₂ as a PPAR--specific ligand. Some commercially available preparations contain bioactive contaminants that may affect functional studies (19). Other studies have shown that 15d-PGJ₂ may affect cellular function and transcriptional process through PPAR--independent pathways (20). Therefore, we confirmed the functional data in these studies using the specific PPAR- agonist ciglitazone. Ciglitazone is a relatively weak activator of PPAR- compared with 15d-PGJ₂ and has an EC₅₀ of about 3 µM. As depicted in Figure 3B, treatment of cells with increasing concentrations of ciglitazone also resulted in a dose-dependent activation of (AOx)₃-TK-luciferase.

To further confirm that the activation of (AOx)₃-TK- luciferase was due solely to PPAR-, similar experiments were performed using this reporter gene to examine the possible role of PPAR--specific activity. In these experiments (Figure 3C), the cells were rested overnight in 0.5% human serum and then stimulated for 24 h with a PPAR--specific ligand, WY14643. In control experiments, WY14643 weakly activated (AOx)₃-TK-luciferase in cells cotransfected with an expression vector for murine PPAR-. However, WY14643 failed to activate (AOx)₃-TK-luciferase in untransfected cells. Together with protein and Northern data, these results indicate that NIH-A549 cells express little if any PPAR-.

Finally, we examined the effects of IL-4 on ciglitazone-induced (AOx)₃-TK-luciferase activity. If IL-4 increases PPAR- expression in NIH-A549 cells, ciglitazone should stimulate higher levels of PPAR--dependent activity in IL-4-treated cells. In these experiments, NIH-A549 cells were transfected with (AOx)₃-TK-luciferase and the normalizing vector and then treated with IL-4 (10 ng/ml). The following morning, ciglitazone was added to the cultures for an additional 24 h. As expected, the highest level of promoter activity was observed in cells treated with IL-4 and ciglitazone (Figure 3D).

Importantly, the studies using PPAR- agonists were completed in the absence of a cotransfected PPAR- expression vector and indicate that the transcriptional activation of (AOx)₃-TK-luciferase in these experiments results from the activation of endogenous PPAR-. Thus, NIH-A549 cells not only express PPAR- but also possess intact signaling pathways for mediating its effects on gene transcription.

PPAR- Activation Blocks Cytokine-Induced iNOS Expression

We next focused on characterizing the immunomodulatory effects of PPAR- in airway epithelial cells. In murine macrophages, PPAR- inhibits cytokine-induced iNOS and gelatinase B expression through nuclear factor (NF)- B, Stat protein, and activator protein (AP)-1-dependent mechanisms. On the basis of these data, we examined the effect of PPAR- activation on the expression of two airway epithelial cytokines that that also depend on NF-B and AP-1: iNOS (21) and IL-8 (22).

Similar to previously published findings, resting NIH-A549 cells did not produce detectable levels of iNOS protein in vitro (Figure 4A). Stimulation of NIH-A549 cells with proinflammatory cytokines dramatically upregulated iNOS protein expression. The cytokine-induced increase in iNOS expression was blocked by both 15d-PDJ₂ and ciglitizone in a dose-dependent manner.

View larger version (37K): [in this window] [in a new window]   Figure 4.   Activation of PPAR- inhibits cytokine-induced iNOS expression in airway epithelial cells. NIH-A549 cells were rested overnight in 0.5% human serum and then treated with inflammatory cytokines (10 ng/ml IL-1, 10 ng/ml TNF-, and 100 u/ml IFN-) alone or the cytokines plus either 15d-PGJ2 or ciglitazone at the indicated concentrations for 24 h. A total of 50 µg total protein was loaded per lane and immunoblotting was performed as described.

Activation of PPAR- Downregulates IL-8 Secretion

NIH-A549 cells secreted little IL-8 at baseline (Figure 5A). Stimulation with inflammatory cytokines significantly upregulated secretion of IL-8 protein as detected by enzyme-linked immunosorbent assay. Both 15d-PGJ₂ and ciglitazone inhibited cytokine-induced IL-8 secretion, although the effects of 15d-PGJ₂ were more profound (Figure 5B). Because IL-4 stimulates PPAR- expression, we postulated that pretreating the cells with IL-4 would increase the PPAR-- dependent transrepression of IL-8 secretion. Indeed, ciglitazone dramatically inhibited the cytokine-induced secretion of IL-8 in IL-4-primed NIH-A549 cells (5B).

View larger version (17K): [in this window] [in a new window]   Figure 5.   Activation of PPAR- inhibits cytokine-induced IL-8 expression in airway epithelial cells. NIH-A549 cells were rested overnight in 0.5% human serum in the presence or absence of IL-4 (10 ng/ml) and then treated with inflammatory cytokines alone or the cytokines plus 15d-PGJ2 (1 and 5 µM) or ciglitazone (5 and 25 µM) for 24 h as indicated. Data are presented as means ± standard error of the mean IL-8 concentration of two independent experiments. Each condition was performed in triplicate. *Significant difference (P < 0.05) in IL-8 secretion between cytokine-stimulated cells and cytokine-stimulated cells in the presence of ciglitazone and IL-4.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Both iNOS and IL-8 are believed to play critical roles in airway host defense. NO synthases are responsible for the in vivo synthesis of NO, a short-lived molecule that is an effective bactericidal agent and may also regulate expression of various proinflammatory genes, such as IL-8 (23). IL-8 is a potent chemoattractant and activator of neutrophils (24). To our knowledge these findings are the first to demonstrate that activation of airway epithelial PPAR-, a ligand-dependent activator of gene transcription, significantly inhibits the cytokine-mediated induction of inflammatory mediators iNOS and IL-8.

Chang and Szabo (25) recently described the expression of PPAR- in a variety of non-small cell lung cancer cell lines. Treatment of multiple cell lines using higher concentrations of ciglitazone and 15dPGJ₂ resulted in the induction of several protein markers of differentiation (25). Whether these effects were mediated strictly through PPAR- activation was unclear given the high doses of ligands used and the differing effect of the ligands on PPAR- activation and expression. In particular, high doses of 15d-PGJ₂ have been found to induce signaling through non-PPAR--dependent pathways (20). In our study, we demonstrate that stimulating airway epithelial PPAR- using either 15d-PGJ₂ or a synthetic ligand specific for PPAR-, ciglitazone, resulted in the transactivation of a PPAR-dependent promoter, (AOx)₃-TK-luciferase. This strongly suggests that the effects of these ligands on inflammatory gene expression are indeed mediated through activation of PPAR-.

Both 15d-PGJ₂ and ciglitazone abrogated the cytokine-induced iNOS expression in airway epithelial cells. Although ciglitazone alone did not significantly affect the upregulation of IL-8 secretion in response to proinflammatory cytokines, it significantly inhibited cytokine-induced IL-8 secretion in cells that had been pretreated with IL-4.

In monocytes and macrophages, PPAR- inhibits the expression of various cytokines partly by preventing the activation of the NF-B, AP-1, and Stat transcription factors (26). Both IL-8 and iNOS expression are regulated primarily by transcriptional mechanisms dependent on the regulatory influences of NF-B, AP-1, and NF-IL-6 response elements. The differential effects of PPAR- activation on iNOS and IL-8 gene expression in airway epithelial cells suggest that there may be important differences in the contribution of each pathway to IL-8 and iNOS gene expression and that there are differences in the effects of PPAR- activation on individual signal transduction pathways. The effects of ciglitazone on IL-8 expression in airway epithelial cells are consistent with a recent report in which PPAR- ligands inhibited IL-1-induced IL-8 expression in colonic epithelial cells (27). In contrast, activation of PPAR- failed to decrease lipopolysaccharide-induced IL-8 secretion from human monocytic THP-1 cells (28). One explanation may be that the levels of endogenous PPAR- in THP-1 cells were insufficient to inhibit IL-8 secretion. Although airway epithelial cells expressed much higher levels of PPAR- than did THP-1 monocyte controls, the inhibitory effect of ciglitazone on IL-8 secretion required that PPAR- levels be upregulated by priming with IL-4. In addition, the use of FCS by the other groups may also have contributed to conflicting results. Our experiments used human serum, which we have found to be important in maintaining IL-4 responsiveness.

The effects of PPAR- on cytokine-induced IL-8 secretion were less profound compared with its effects on iNOS expression. PPAR- activation using ciglitazone alone produced little if any change in IL-8 secretion. However, ciglitazone is a relatively weak activator of PPAR-. Ciglitazone had a much greater effect on IL-8 secretion in IL-4- primed cells, consistent with our finding that IL-4 upregulates PPAR- expression. Although these results do not rule out the possibility that ciglitazone may be working through non-PPAR--dependent pathways in IL-4-treated cells, they do suggest that the effects of PPAR- activation are more potent in the presence of IL-4. IL-4 also upregulates 12/15-LO in NIH-A549 cells (29). Considering that human 12/15-LO generates two known PPAR- ligands, 13S-HODE and 15S-HETE, it suggests that IL-4 can coordinate the expression of PPAR- and an enzyme that generates functional ligands in NIH-A549 cells as it does in murine macrophages (13).

The 12/15-LO and its metabolites have long been associated with anti-inflammatory properties. The activation of PPAR- suggests yet another mechanism in addition to lipoxin synthesis, altered phosphatidyl inositol/protein kinase C signaling and 5-lipoxygenase pathway antagonism through which 12/15-LO can downregulate certain proinflammatory responses. Recently, lipoxin A₄, a 12/15-LO metabolite, was shown to inhibit Salmonella typhimurium- induced IL-8 secretion from T84 cells (30). Similarly, overexpression of human 12/15-LO in rat mesangial cells in a rat model of glomerulonephritis, was associated with increased lipoxin A₄ secretion and improved glomerular function (31). Although the role of PPAR- in these studies was not determined, studies linking the IL-4-dependent inhibition of iNOS in NIH-A549 cells may prove to be dependent on the expression of 12/15-LO and PPAR- (32).

These findings may be especially relevant in airway diseases such as asthma. Neutrophils may augment or perpetuate airway inflammation and injury during asthma exacerbations. Studies also suggest that an inflammatory process dominated by neutrophils rather than eosinophils characterizes the airway of patients with severe disease (33, 34). The cause of the airway neutrophilia found in asthma is unclear but may be due to phenotypic differences in mediator synthesis and release between asthmatics and nonasthmatics (35, 36). Elevated levels of IL-8 are found in bronchoalveolar lavage fluid from intrinsic asthmatics (37). Our findings that PPAR- is expressed and upregulated by IL-4 in airway epithelial cells and that activation of airway epithelial PPAR- downregulates expression of inflammatory mediators suggests that PPAR- can act as an anti-inflammatory agent, possibly through 12/15-LO-dependent pathways.

Persistent airway wall inflammation is believed to play a critical role in the development and progression of diseases such as asthma (38). However, therapeutic options aimed at reducing or suppressing airway inflammation are limited. There is growing interest in the use of PPAR- agonists as anti-inflammatory therapies. The antidiabetic thiazolidinediones have been demonstrated to have a protective effect in animal models of atherosclerosis (39) and chronic bowel inflammation (27). Further investigation into the mechanisms through which PPAR- inhibits inflammation should greatly expand our knowledge of the immunomodulatory role of the airway epithelium and also promote the development of novel anti-inflammatory agents for the airways.