Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The acute phase response (APR) is the host's attempt to limit disturbances in homeostasis. Hallmarks of this inflammatory reaction include fever, leukocytosis, increased secretion of glucocorticoids (GCs), and alterations in the metabolism and production rates of certain plasma proteins. Infection, injury, immunologic disorders, and neoplasia initiate the APR by generating a local reaction that results in the release of the first wave of the proinflammatory cytokines from activated tissue macrophages and blood monocytes. At the site of injury, interleukin (IL)-1 and tumor necrosis factor (TNF) target stromal cells, such as fibroblast and endothelial cells, which generate a secondary wave of cytokines, including IL-6. Distally, IL-1, IL-6, and GCs modulate the production and secretion of a group of plasma proteins, known as the acute phase proteins (APPs), from the liver, a primary target of the APR. The APPs play an important role in controlling tissue damage and promoting healing (1).

Fibrinogen (FBG) is a major plasma protein produced constitutively by the liver. It is composed of three polypeptide subunits (A, B, and ), and each is the product of a single-copy gene (6). FBG is the final component in the coagulation cascade where thrombin converts it into fibrin. The fibrin monomer polymerizes to form the insoluble gel that arrests hemorrhaging at sites of tissue damage and initiates wound repair (6). During the APR, the expression of the FBG genes is coordinately upregulated, resulting in a 2- to 20-fold increase of the plasma levels (3, 7). IL-6 and GCs are the primary mediators of increased FBG expression during the APR. Although the exact mechanisms are not fully understood, transcriptional regulation seems to be an important element of control for FBG expression (8). In contrast, IL-1 seems to have little effect or inhibits the hepatic expression of FBG (1, 2, 9- 11). Despite this, there are few studies that examine the molecular mechanisms of IL-1-mediated regulation of FBG production.

The role of proinflammatory cytokines in lung pathophysiology is an area of great interest because the lung is the target of various inflammatory diseases, ranging from asthma to pneumonia (12). Further, lung cells act as both effectors and targets of these inflammatory mediators. Our previous studies demonstrated the induction of FBG expression by the APR mediators IL-6 and GC in an adenocarcinoma cell line (A549) that is derived from human alveolar type II epithelial cells (16). In addition, FBG is produced by the lung epithelium of Pneumocystis carinii-infected ferrets and mice (17). Moreover, the expression of haptoglobin, an APP that displays antioxidant and antimicrobial activities, is upregulated in alveolar epithelial type II cells during inflammation (18). Other studies indicate that P. carinii infection can result in hepatic APP production. For example, elevated levels of plasma C-reactive protein, an APP, were detected in patients who died of P. carinii pneumonia (PCP) secondary to acquired immunodeficiency syndrome (AIDS) (19). A second study examined cases of extrapulmonary P. carinii infection in patients with AIDS; seven out of 37 individuals had hypoalbuminemia, another indicator of an APR (20). Together, these data indicate that the inflammation produced in the lung by infection generates a local as well as a systemic APR.

In the present study, human liver (HepG2) and lung (A549) epithelial cells were used as in vitro models of systemic and local inflammation, respectively. The production of FBG was measured at the level of messenger RNA (mRNA) and protein after treatment with mediators of the APR: IL-6 and IL-1 plus dexamethasone (DEX), a synthetic GC. The results demonstrated that IL-1-mediated downregulation of hepatic and pulmonary FBG gene expression occurs by similar mechanisms. More interestingly, however, there was an unexpected difference in the response of the two cell types to treatment with IL-1 plus DEX. Although the production of FBG mRNA and protein remained downregulated in the liver cells, it was increased above control levels in the alveolar epithelial cell. This cell type-specific stimulation of FBG production by GC in the presence of IL-1 may play a role in pulmonary wound repair mechanisms during a local APR.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cells and Culture Conditions

A549 human lung epithelial carcinoma cells (ATCC-CCL 185) were maintained in Kaighn's F12 media (Irvine Scientific, Santa Ana, CA) supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 10% fetal bovine serum (FBS). HepG2 human hepatoma cells (ATCC-HB 8065) were maintained in Eagle's minimum essential media (GIBCO BRL, Rockville, MD) with 2 mM L-glutamine, 100 U/ml penicillin, 0.1 mg/ml streptomycin, 0.1 mM nonessential amino acids, 1.0 mM sodium pyruvate, 15 mM tricine (pH 7.4), and 10% FBS. Cells were seeded into six-well plates, grown to  90% confluency, then treated with cytokines and/or DEX for 24 h at 37°C. Except for the dose-dependency studies, the treatments consisted of fresh medium containing one or more of the following: 25 U/ml recombinant human IL-6 (GIBCO BRL, Minneapolis, MN) (16), 500 U/ml recombinant human IL-1 (R&D Systems) (1, 2, 10, 11), or 0.1 µM DEX (16). For the cycloheximide (CHX) studies, cells were pretreated with fresh medium containing 5 µg/ml of CHX for 2 h, then the appropriate mediators were added to each well for an additional 24 h (2). To examine the effects of cytokines and DEX treatment on FBG mRNA stability, cells were treated with the appropriate mediators for 24 h and washed with warm medium, then new medium with fresh mediators and 5 µg/ml of actinomycin D (ActD) was added to each well for the indicated times (21).

RNA Isolation and Northern Blot Analysis

Total RNA was isolated using TriReagent (Molecular Research Center, Inc., Cincinnati, OH) according to manufacturer's instructions. Northern blot analysis was performed as previously described (22). Briefly, total RNA from each well was denatured in a glyoxal/dimethyl sulfoxide denaturing mix, then electrophoresed on a 1.2 or 1.4% agarose gel in 10 mM sodium phosphate buffer (pH 7.0) with recirculation. The RNA was transblotted to Zeta-Probe membrane (Bio-Rad, Hercules, CA) in 0.5× Tris acetate-ethylenediaminetetraacetic acid (EDTA), air-dried, and fixed by baking at 80°C in vacuo for 1 to 2 h. The fixed blots were prehybridized at 65°C in 0.5 M Na₂HPO₄ (pH 7.2), 1 mM EDTA, and 7% sodium dodecyl sulfate (SDS) for  1 h.

Restriction endonuclease digestions were performed to release the inserts of the following probes: human FBG 1,000 base pairs (bp), SacI and HindIII; human -actin 2,000 bp, BamHI. The inserts were purified by gel electrophoresis and recovered using the Sephaglas BP kit (Pharmacia Biotech, Piscataway, NJ) according to manufacturer's instructions. Radioactive probes were labeled with [-³²P]deoxycytidine triphosphate (dCTP) (DuPont NEN, Boston, MA) by the random primer method (GIBCO BRL). The radiolabeled probe was denatured by boiling and added to the prehybridization buffer so that the final activity was about 1 × 10⁶ cpm/ml. Hybridization occurred at 65°C for 16 to 24 h. The blots were washed in Wash Solution #1 (40 mM Na₂HPO₄ [pH 7.2], 1 mM EDTA, and 5% SDS) for 60 min at 65°C, followed by three washes in Wash Solution #2 (40 mM Na₂HPO₄ [pH 7.2], 1 mM EDTA, and 1% SDS) at 65°C for 30 min each. After air-drying, the blots were exposed to X-ray film. The blots were stripped for reprobing by incubation with Wash Solution #1 at 95°C for 60 min, then three incubations with Wash Solution #2 at 95°C for 30 min each. The blots were exposed to X-ray film overnight to confirm the absence of signal, then hybridized with a different probe.

The results of the Northern blots were analyzed by densitometry scanning using the NIH Image 1.59 program. Except in the ActD studies, the abundance of the FBG signal was normalized to the -actin signal to account for differences in the amount of total RNA loaded. The fold-induction of FBG mRNA abundance was calculated relative to the appropriate control as indicated for each experiment. P values were obtained using five-way analysis of variance (ANOVA) for the data set collected and shown in Figure 1 (see RESULTS). This statistical analysis was used to measure the effects of multiple factors in various combinations (i.e., experiment number, cell type, IL-6, IL-1, and DEX). The five-way ANOVA separates out the specific effects as well as determining the influence of one factor on another. This method also reduced the chance of error inflation by determining whether there were any differences among the means of any of the groups. The two-tailed Student's t test was used when examining only one factor (e.g., CHX) on the system. A P value < 0.05 indicated statistical significance.

View larger version (18K): [in this window] [in a new window]   View larger version (52K): [in this window] [in a new window]   View larger version (23K): [in this window] [in a new window]   Figure 1.   The effects of APR proinflammatory cytokines and DEX on FBG mRNA abundance in A549 and HepG2 cells. Total RNA was isolated using TriReagent and analyzed by Northern blotting. The abundance of FBG mRNA was normalized relative to -actin levels. The fold-induction was calculated relative to the control value for each cell line. Data in the graphs (A and C) are expressed as the mean (± SEM) of five to seven separate experiments for each treatment condition. Statistical analysis was by five-way ANOVA. Representative Northern blots of A549 and HepG2 cells were treated with the indicated mediators (B and D).

Metabolic Labeling and Immunoprecipitation

FBG protein was metabolically labeled using 40 µCi/ml of [³⁵S]methionine and cysteine Express Protein Labeling mix (Dupont NEN) for 24 h in the presence of various combinations of IL-6, IL-1, and DEX. Rabbit antihuman FBG antibody (DAKO, Carpenteria, CA) bound to Protein A-Sepharose beads (Pharmacia Biotech) was added directly to the culture supernatant of each sample to immunoprecipitate nascent FBG. Each sample was resolved by SDS- polyacrylamide gel electrophoresis (PAGE) under reducing conditions and analyzed by fluorography (23).

Nuclear Run-On Transcription

HepG2 and A549 nuclei were isolated from approximately 1 to 2 × 10⁷ cells; the run-on transcription was performed as previously described (16, 24) with modifications. Equal volumes of nuclei and 2× reaction buffer (2× = 10 mM Tris-HCl [pH 8.0]; 5 mM MgCl₂; 0.3 M KCl; 4 mM MnCl₂; 1 mM dithiothreitol; 0.2 mM EDTA; 20 U/ml RNasin; and 4 mM each of adenosine triphosphate, guanosine triphosphate, and CTP) were mixed with 250 µCi [³²P]uridine triphosphate (3,000 Ci/mmol, 10 µCi/µl; Dupont NEN) and incubated at 30°C for 30 min with constant gentle shaking to allow elongation of nascent mRNA. DNA and protein were degraded by treatment with DNase I and Proteinase K. The labeled transcripts were isolated by using TriReagent LS according to the manufacturer's instructions and purified further using Sephadex G-50 spin columns (Pharmacia Biotech). Scintillation counting was used to determine the specific activity of each sample, which was at least 6 × 10⁶ cpm/ml.

Slot blots were made containing 5 µg of linearized, denatured plasmid DNA: FBG, -actin, or pGEM7f() plasmid DNAs (described previously for the Northern blotting method). The blots were hybridized with all of the [³²P]-labeled nascent transcripts in Northern hybridization buffer (see Northern protocol) for 36 h at 65°C. After hybridization, the blots were washed twice in Northern Wash Buffer #1 for 30 min at 65°C and three times in Northern Wash Buffer #2 for 30 min at 65°C. The air-dried blots were exposed to X-ray film. The -actin signal was used to determine the relative rates of transcription.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

IL-6 and IL-1 Similarly Modulate FBG Gene Expression in A549 and HepG2 Cells

Human lung adenocarcinoma (A549) and hepatoma (HepG2) cell lines were used to elucidate any differences between the in vitro effects of proinflammatory cytokines on FBG mRNA expression in the lung and liver cells. These cell lines were treated with IL-6 and/or IL-1 for 24 h and the abundance of FBG mRNA was analyzed by Northern blotting. The fold-induction was calculated relative to the control values for each cell type. Data represent averages ± the standard error of the mean (SEM) of five to seven experiments per condition (Figure 1A) with a representative Northern blot shown (Figure 1B). The results indicate that the effects of APR cytokines were similar in both liver and lung epithelial cells. Treatment with IL-6 increased the steady-state abundance of FBG mRNA 2-fold in A549 cells and 1.3-fold in the HepG2 cells in a statistically significant manner (P = 0.03) as determined by five-way ANOVA. In contrast, IL-1 reduced FBG abundance to about 50% of control values in a statistically significant manner (P < 0.0001), even in the presence of IL-6. Thus, the cell type was not a factor in the IL-1- mediated inhibition of FBG gene expression.

Lung Cell-Specific Modulation of FBG Gene Expression by DEX and IL-1

The other APR mediator that has a significant role in the regulation of FBG expression is GC; therefore, the effect of DEX on FBG expression in the lung and liver cell lines was studied. Again, Northern blot analysis was used to assess the effect of DEX in the absence and presence of cytokines on the abundance of FBG mRNA in A549 and HepG2 cells. The data are shown as a compilation of five to seven experiments per condition (Figure 1C) and by a representative Northern blot (Figure 1D). The results of these experiments revealed that the addition of DEX regulated the expression of FBG mRNA differently in the two cell lines. The abundance of FBG mRNA was elevated 1.8 ± 0.7-fold due to DEX treatment in the A549 cells, whereas DEX treatment did not effectively alter FBG transcripts in the HepG2 cells (0.9 ± 0.1). Statistical analysis by five-way ANOVA indicated that the effect of DEX alone on FBG mRNA in A549 cells was cell type- specific and statistically significant (P = 0.0003). The FBG mRNA level remained inhibited in HepG2 cells (~ 50% below control) in the presence of IL-1 + DEX; however, IL-1 + DEX treatment produced opposite results in the A549 cells. In the presence of IL-1, DEX treatment of A549 cells resulted in a > 2.5-fold increase in FBG mRNA abundance over control levels in a cell type-specific and statistically significant manner (P = 0.045), an effect that did not change in the presence of IL-6. Thus, DEX increased FBG mRNA abundance in A549 cells in a cell type-specific manner when used alone or in combination with IL-1.

DEX Rescue Is Dependent on DEX and IL- Concentrations

To determine whether the induction or inhibition of FBG mRNA abundance was dependent on the concentration of the mediator used, Northen blot analysis was performed and the relative abundance of FBG mRNA quantitated (Figure 2). The results indicate that FBG expression in A549 cells was induced by IL-6 in a dose-dependent manner with 0.1 µM DEX enhancing FBG induction at low doses of IL-6 (25 to 250 U/ml). In the absence of DEX, it would require 600 U/ml to equal the 2.7-fold increase in FBG abundance observed with 250 U/ml IL-6 plus 0.1 µM DEX (Figure 2A). Thus, IL-6 induction of FBG gene expression in A549 cells is enhanced by DEX as previously shown to occur in HepG2 cells (1, 2, 10, 11). The inhibition of FBG mRNA abundance by IL-1 was also shown to be dose-dependent; maximal inhibition was achieved by 500 U/ml IL-1 (Figure 2B). In the presence of 0.1 µM DEX, the abundance of FBG mRNA was "rescued" over a range of IL-1 concentrations (Figure 2B). Further, rescue of FBG mRNA abundance in the presence of 500 U/ml IL-1 was dependent on DEX over a concentration range of 0 to 1 µM. The effect of increasing DEX concentrations on rescuing FBG abundance was maximal at 0.1 µM, the concentration used throughout these studies (Figure 2C).

View larger version (15K): [in this window] [in a new window]   Figure 2.   Effects of increasing doses of APR mediators on FBG gene expression. Relative steady-state levels of FBG mRNA were measured after normalization to -actin levels by Northern blotting and densitometric scanning of autoradiographs. Induction of FBG gene expression occurred in the presence of increasing doses of IL-6 with and without DEX (A). Increasing doses of IL-1 inhibited FBG, whereas addition of 0.1 µM DEX "rescued" the inhibitory effect of IL-1 on FBG gene expression (B). In the presence of a constant amount of IL-1 (500 U/ml), DEX rescue of FBG mRNA abundance occurred in a dose-dependent manner (C).

DEX Rescue of IL-1 Inhibition of FBG Protein Production Is Cell Type-Specific

The pattern of FBG production in the A549 and HepG2 cells due to IL-6, IL-1, and/or DEX was analyzed at the protein level by SDS-PAGE of immunoprecipitated FBG that was metabolically labeled (Figure 3). The cells were treated with the APR mediators as previously described for the Northern blots with the addition of [³⁵]S-methionine and cysteine for 24 h. The relative amount of FBG protein produced by the various treatments compared with the untreated control was determined by densitometric scanning of the fluorographs (n = 3). In the absence of IL-6, IL-1 showed minimal effect on inhibition of protein production in HepG2 (0.77 ± 0.24) and A549 (1.0 ± 0.1) cells; also, in the presence of IL-6, IL-1 inhibition of FBG protein production in both cell types was less (HepG2, 0.81 ± 0.18; A549, 0.86 ± 0.22) than observed at the level of mRNA (Figure 1). DEX rescue of the IL-1-mediated inhibition of FBG was apparent only in the A549 (1.94 ± 0.35 SEM) and not in the HepG2 (0.70 ± 0.22) cells. Although DEX treatment caused a 1.8-fold increase in FBG mRNA abundance in A549 cells (Figure 1) due to enhanced stability (see subsequent discussion regarding Figure 6), this effect did not result in an increase in the amount of FBG protein (A549, 1.1 ± 0.09; HepG2, 0.9 ± 0.15) (Figure 3).

View larger version (59K): [in this window] [in a new window]   Figure 3.   SDS-PAGE of metabolically labeled, immunoprecipitated FBG secreted from A549 and HepG2 cells treated with APR cytokines and DEX. A549 (left half ) and HepG2 (right half ) cells treated with IL-6 and IL-1 (with and without DEX) were metabolically labeled with [35S]cysteine plus 35S-methionine for 24 h. FBG from the conditioned media immunoprecipitated with anti-FBG antibodies was separated under reducing conditions by SDS-PAGE. [14C]-labeled relative moleculear mass (Mr) markers (middle lane) in kiloDaltons are as follows, from top to bottom: 200, 97 (runs as doublet), 68, and 46. The A, B, and  chains of FBG are indicated. This gel is representative of three independent experiments.

View larger version (33K): [in this window] [in a new window]   Figure 6.   The effects of IL-1 and DEX on the stability of FBG mRNA in A549 cells. A549 cells were treated with nothing (control) or with DEX, IL-1, or IL-1 + DEX. After 24 h, the media were replaced with fresh media containing the appropriate mediators and ActD (5 µg/ml). The t1/2 of FBG mRNA was calculated from the average value determined from linear regression analysis of four independent experiments; the Northern blotting from one of these experiments is shown.

DEX Rescue of FBG Gene Expression Requires De Novo Protein Synthesis

To determine whether the actions of these mediators required new protein synthesis, A549 cells were treated with CHX to prevent de novo protein synthesis. The cells were either left untreated or pretreated with CHX; then the cytokines were added for 24 h and the abundance of FBG mRNA was measured by Northern hybridization. The fold-induction of FBG mRNA was calculated relative to the control sample without CHX. The data represent the average ± SEM of four experiments per condition (Figure 4A) with a representative Northern blot shown (Figure 4B). Neither IL-6-mediated induction nor IL-1-mediated inhibition of FBG expression required new protein synthesis because there was little difference (P > 0.05) between the CHX-treated and untreated A549 cells (Figures 4A and 4B).

View larger version (18K): [in this window] [in a new window]   View larger version (42K): [in this window] [in a new window]   View larger version (27K): [in this window] [in a new window]   Figure 4.   The effects of CHX on the expression of FBG mRNA in A549 treated with APR cytokines and DEX. A549 cells were pretreated with 5 µg/ml of CHX for 2 h, then cytokines with and without DEX as indicated in each panel were added to the media for an additional 24 h. The abundance of FBG mRNA was analyzed by Northern blotting. The fold-induction was calculated relative to the control sample without CHX. Statistical analysis was done using a two-tailed Student's t test. Data in the graphs (A and C) are means ± SEM (n = 4). Representative Northern blots of A549 cells treated as indicated are shown (B and D).

To determine whether new protein synthesis was necessary for DEX rescue, A549 cells were treated with CHX in the presence of IL-6, IL-1, and DEX (Figures 4C and 4D). Neither DEX-mediated and IL-6 + DEX-mediated induction nor IL-1-mediated inhibition of FBG gene expression required new protein synthesis (P > 0.05). In contrast, DEX rescue of the IL-1-mediated inhibition of FBG expression in either the presence (P = 0.015) or absence (P = 0.002) of IL-6 required new protein synthesis, suggesting that DEX treatment induces de novo protein synthesis of an inhibitor of IL--mediated downregulation of FBG.

Inhibition of FBG Gene Expression by IL-1 Occurs at the Level of Transcription

Nuclear run-on assays were performed to directly measure the rate of nascent FBG mRNA expression. The amount of FBG gene transcription was evaluated in control A549 cells and cells treated with IL-1 and IL-1 + DEX for 24 h (Figure 5). As positive controls for FBG gene expression, both A549 (not shown) and HepG2 cells were treated with IL-6 + DEX, which upregulated FBG transcription in both cell types as previously shown (see Simpson-Haidaris [16], Figure 3). There was minimal new transcription of the FBG mRNA in the untreated control A549 cells. In addition, after IL-1 treatment, there was no detectable expression of any nascent FBG transcripts (Figure 5, top), suggesting that IL-1 treatment reduced the already low basal level of FBG transcription in the lung epithelial cells. Further, expression of the FBG gene was not upregulated at the level of transcription by treatment with IL-1 + DEX (Figure 5, middle) or DEX (not shown), despite an increase in steady-state abundance as measured by Northern blot analyses (Figures 1C and 1D). These results indicate that there was no new transcription at 24 h due to DEX or IL-1 + DEX treatment.

View larger version (50K): [in this window] [in a new window]   Figure 5.   Nuclear run-on assay of A549 cells treated with IL-1 or IL-1 + DEX. Both HepG2 and A549 cell lines were treated with the indicated mediators for 24 h. IL-6 + DEX treatment of HepG2 cells was done as a positive control for FBG gene induction. The cell nuclei were isolated and nascent mRNA was elongated in the presence of [-32P]uridine triphosphate in a run-on transcription assay. The resulting labeled mRNA probe was used to hybridize to an excess of denatured FBG, -actin, or pGEM7Zf() (negative) plasmid DNA. Data shown are representative of data from three to five independent experiments per condition.

DEX Treatment Alone Enhances the Stability of FBG mRNA in A549 Cells

To measure the effect of IL-1 and DEX on the stability of FBG mRNA in the lung epithelial cell line, the cells were either untreated (control) or treated with IL-1, DEX, or IL-1 + DEX for 24 h. At this time (designated 0 h time point after ActD addition), the old medium was replaced with fresh medium containing ActD and the same concentrations of IL-1, DEX, or IL-1 + DEX. Northern blotting was performed to determine the amount of FBG mRNA remaining after ActD treatment. The half-life (t_1/2) was calculated by linear regression analysis of the relative abundance of FBG mRNA remaining at each time point compared with the relative amount of the 0-h time point from three to five independent experiments per time point (Figure 6A). The t_1/2 of FBG mRNA in the unstimulated A549 cells was about 25 h. DEX alone increased the stability of the FBG message because the t_1/2 was over 30 h. The t_1/2 of FBG mRNA in A549 cells treated with IL-1 with and without DEX were about the same as that of the control, despite the fact that the abundance of the FBG mRNA in cells treated with IL-1 was approximately 50% of control values and 10 to 20% of FBG mRNA levels in the DEX-treated cells (Figure 6). Further, the DEX-enhanced stability of FBG mRNA was not observed in the presence of IL-1.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The mechanisms regulating the constitutive and inducible expression of each of the FBG genes in hepatic tissue have not been fully elucidated; however, numerous reports describe a complex interplay of cis-acting elements and trans-acting factors (25). The available data clearly indicate that the inducible regulation of FBG involves both IL-6, the primary cytokine mediator of APP production, and GC (7, 16, 31, 32). IL-1 is also important in mediating the upregulation of other APPs, such as serum amyloid A, and functions synergistically with IL-6 (33). In contrast, IL-1 downregulates the IL-6-induced elevation of hepatic FBG expression during the APR both in vivo and in vitro (1, 2, 9, 34).

We have demonstrated elevated levels of FBG mRNA in lung epithelium of P. carinii-infected SCID mice and immunosuppressed ferrets (17). In addition, lung alveolar epithelial cells (A549) produce FBG when induced with IL-6 and DEX (16). One of the host responses to pulmonary infection by P. carinii is the increased production of various cytokines. Wright and colleagues (14) demonstrated elevated levels of IL-1 and IL-6 mRNA in lung cells of P. carinii-infected SCID mice that were immunologically reconstituted. Thus, the proinflammatory mediators that modulate FBG gene expression are expressed locally by lung cells. Further, it has been observed that AIDS patients with PCP mount less severe pulmonary inflammation than do non-AIDS patients with PCP, such as those undergoing chemotherapy or bone-marrow transplantation, whose immune competency is restored following treatment (35). Consistent with this observation, the intense localized pulmonary inflammation in the reconstituted SCID mice that are resolving the P. carinii infection corresponds to areas of increased IL-1 and TNF- expression. Unfortunately, this host inflammatory response may exacerbate the course of the disease (14).

Paradoxically, GC treatment is a recommended adjunctive anti-inflammatory therapy for PCP, a disease that occurs in already immunocompromised and immunosuppressed patients. It is thought that GCs attenuate the host inflammatory response to the infection and improve lung function (36). The positive effects of DEX treatment to ameliorate the host inflammatory response during PCP likely include improvement in surfactant homeostasis (36), increased lung mechanics by decreasing vascular permeability (37), and inhibition of the synthesis of proinflammatory cytokines (38). Although the exact role of newly synthesized FBG in the lung is not known, it is likely involved in wound repair mechanisms as a component of the provisional extracellular matrix. We know that FBG secreted basolaterally from A549 cells is incorporated into the extracellular matrix as conformationally altered, insoluble FBG (23, 39). Further, the intra-alveolar FBG found in P. carinii-infected lungs appears to aggregate organisms at the apical face of the type I epithelium (17). Because IL-6, IL-1, and GC influence the course of P. carinii infection, pneumonitis, and resolution of infection, we analyzed the mechanisms by which IL-6, IL-1, and DEX modulate FBG gene expression using alveolar (A549) and hepatic (HepG2) epithelial cell lines as models.

Both HepG2 and A549 cells responded to treatment with IL-6 with and without DEX in a similar manner, as shown by the upregulation of FBG gene expression and the assembly and secretion of intact FBG. Further, IL-1 downregulated FBG mRNA and, to some extent, protein production in both lung and liver epithelial cells (Figures 1 and 3) in a dose-dependent manner in A549 cells (Figure 2). In addition, neither induction by IL-6 with and without DEX and DEX alone, nor downregulation by IL-1 of FBG expression, required new protein synthesis (Figure 4). Together, these data are in agreement with previously published studies on both rat and human hepatoma cell lines and rat primary hepatic cell cultures, which show that IL-1 has a minimal inhibitory rather than stimulatory effect on FBG production (1, 2, 9, 34). The inhibition of protein production by IL-1 is consistently less than that oberved at the mRNA level in our study, as well as in the previously cited studies. Although the exact mechanism for this is unclear, the fact that pre-existing intracellular pools of A and  chain polypeptides are found in both HepG2 (40) and A549 cells (our unpublished observations) suggests that the time course for IL-1 inhibition of protein production may be delayed. Further, the direct contribution of IL-6, GC, or IL-1 on the regulation of FBG gene expression and protein production in vivo by both lung and liver epithelium is more difficult to interpret. Previously it was believed that IL-1 induced FBG gene expression (44); however, this effect was later shown to be due to IL-1 induction of IL-6. Because the FBG genes are single-copy, it is likely that the same signal transduction pathways are activated by these APR mediators in both lung and liver epithelial cells to modulate FBG expression. Although FBG gene expression is upregulated in lung epithelium in vivo during PCP (17), the regulation of the FBG gene in the A549 cell line may not be exactly analogous to its regulation in vivo.

The nuclear run-on studies confirmed that IL-1 inhibition of FBG mRNA expression occurred at the level of transcription (Figure 5). In contrast, the elevated abundance of FBG mRNA at 24 h due to IL-1 + DEX treatment ("DEX rescue") was not mediated directly by increased transcription of the gene (Figure 5). In addition, de novo protein synthesis was not required for regulation of the FBG gene mediated by IL-6, DEX, IL-6 + DEX, or IL-1 (Figure 6), suggesting that pre-existing factors function to induce or inhibit transcription by these mediators. This is consistent with the concept that IL-6 regulates the hepatic APR by activating latent transcription factors of the signal transducer and activator of transcription (STAT) 1 and 3 via the Janus family of kinases (45). On the other hand, IL-1 relays its signal to the nucleus via nuclear factor (NF)-B, another pre-existing transcription factor. NF-B plays a critical role in the regulation of the inflammatory response, immune system, stress response, apoptosis, and viral replication (46, 47). Interestingly, binding sites for STAT3 and NF-B can be found on promoters of several APP genes. Both STAT3 and NF-B are capable of binding a DNA motif derived from the 2-macroglobulin promoter, which contains overlapping STAT3 and NF-B elements. This study suggests that NF-B and STAT3 can regulate each other's functions through competition for the overlapping DNA binding sites (48). STAT3 may play a role in regulation of the rat FBG chain gene by IL-6 (49); however, the trans-acting factor(s) that mediate(s) IL-6 upregulation of the human FBG gene have not been definitively identified. As a result, it is attractive to hypothesize that IL-1 downregulates FBG gene transcription by activating NF-B, which can compete with the IL-6-inducible and constitutively expressed transcription factors that bind to the FBG promoter region.

The striking observation from this study is the cell type-specific regulation of FBG expression by DEX in the presence of IL-1, a phenomenon we termed "DEX rescue." It is well established that DEX enhances IL-6- mediated FBG gene expression in hepatocytes (2, 7, 29, 32, 50), and more recently, in extrahepatic epithelial cells (see Figure 2A) (16, 31). However, DEX treatment has never been shown to rescue the expression of FBG genes in hepatocytes treated with IL-1. The studies using ActD suggest that the increase in mRNA stability due to the DEX treatment alone cannot account for the DEX rescue of FBG abundance in the presence of IL-1 (Figure 6). Further, this increase in mRNA stability did not correspond with an increase in intact FBG protein consisting of A, B, and  chain, likely due to other factors regulating the coordinated transcription of the A and B chain genes (16). Instead, we hypothesize that DEX rescue is due to DEX-mediated induction of new synthesis of an intermediate, unrelated gene product that counteracts the inhibition of FBG gene expression and protein production by IL-1. Potential mediators of DEX rescue need to be inducible by DEX and inhibit the IL-1 signal transduction pathway. In vitro studies demonstrated that the IL-1 receptor antagonist (IL-1Ra) restored the downregulation of IL-6-induced FBG production by IL-1 in hepatoma cells (9). IL-1Ra functions as a naturally occurring antagonist of IL-1, and is induced by DEX in bronchial epithelial cells (51). Due to the requirement for new protein synthesis in DEX rescue, it is possible that DEX induction of IL-1Ra in the lung A549 cells is a potential mechanism. However, inasmuch as it has been demonstrated that liver tissue and HepG2 cells produce IL-1Ra (52), it is unlikely that this is the major mechanism regulating the cell type-specific DEX rescue of FBG production by A549 cells.

Alternative mechanisms for DEX rescue include the de novo synthesis of the NF-B inhibitor IB due to DEX treatment. GCs induce the synthesis of new IB, which binds NF-B and inactivates it (53, 54). Recently, Zhang and associates (49) showed that IL-6-activated STAT3 acts as a coactivator for ligand-activated glucocorticoid receptor via the GC response element (GRE) to augment GC signaling in the absence of a STAT3 DNA binding motif. Conversely, activated GCs can interact with the IL-6 response element via STAT3 without a GRE (49). Currently, we are experimentally investigating the likelihood of this mechanism in promoting IL-1 downregulation of FBG gene expression in lung and liver epithelial cells. The complexity of how GCs promote anti-inflammatory activity mechanistically is beginning to be understood (53- 55). In the meantime, it is very clear that unraveling the mechanisms of both IL-1-mediated downregulation and the DEX rescue of FBG gene expression and FBG protein production will be complex, but a better grasp of these processes will increase our understanding of inflammation, particularly in the lung.