Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The most common lethal autosomal recessive disease of the human Caucasian population is cystic fibrosis (CF) (1), with an estimated incidence of 1:2,500. The basic abnormality in CF is a defect of cyclic adenosine monophosphate (cAMP)-regulated chloride secretion by epithelial cells (2). The CF transmembrane conductance regulator (CFTR) gene, located on chromosome 7 at position 7q31, encodes a cAMP-dependent chloride channel in the apical membrane of epithelial cells of the airway and other tissues (5). Following the cloning and sequencing of the CFTR gene in 1989 (2, 6), correction of the basic genetic defect in airway cells of patients with CF by somatic cell gene therapy became feasible. A number of gene delivery systems for this purpose are currently under consideration, including the use of recombinant adenovirus vectors. We evaluated the utilization of a replication-deficient recombinant adenovirus as a viral vector for delivering normal CFTR complementary DNA (cDNA) to airway epithelial cells, with the goal of expressing this gene and correcting the chloride channel defect in the respiratory tract of individuals with CF (7).

A recombinant adenovirus with deletions in the E1 and E3 regions of the viral genome is efficient as a vector in transferring CFTR cDNA into airway epithelial cells in vitro (8). This replication-deficient adenovirus vector is particularly attractive because it has a natural tropism for airway epithelial cells (13). Introduction of this replication-deficient recombinant adenovirus into the lungs of animals resulted in a nonspecific inflammatory response (14). The instillation of a recombinant adenovirus vector containing the Rous sarcoma virus long-terminal-repeat promoter, which drove either a nuclear-targeted gene (AV1LacZ4) or CFTR cDNA (Av1Cf2) in the cotton rat lung, resulted in neutrophil-dominated alveolar and peribronchial cellular infiltration at 3 d after administration (18). The nature of the infiltrate changed from neutrophil-dominated during the first few days to macrophage- and mononuclear cell-dominated by Day 10. Instillation of the same vector into the lungs of macaque monkeys was also associated with a mild to moderate inflammation, the intensity of which was dose-dependent (15). A related recombinant adenovirus vector, utilizing a cytomegalovirus (CMV) promoter and expressing the -galactosidase gene (LacZ), induced a moderate inflammation predominantly in the distal lung parenchyma of the baboon (16). Nonhuman primates exposed to an adenovirus vector developed an inflammatory response at 3 d after exposure. Analysis of bronchoalveolar lavage fluid (BALF) for interleukin (IL)-8 showed increased levels of this cytokine in animals that received a high dose of the vector, suggesting that IL-8 might play a role in the vector-induced inflammation seen in these animals (15). We have shown that in vitro cellular transduction of A549 bronchial epithelial cells by Av1LacZ4, at a multiplicity of infection (MOI) of 50 virus particles per cell, resulted in gene delivery to 80 ± 5% of the cell monolayer. Furthermore, by 24 h, IL-8 messenger RNA (mRNA) transcript and neutrophil chemoattractant activity in supernatants from Av1LacZ4-transduced cells were significantly higher than in supernatants from uninfected control cells. IL-8 mRNA and protein levels remained increased for 96 h (19). These findings show that gene delivery to the airway epithelium with the serotype 5 adenovirus (Ad5)-based expression vector results in IL-8 gene activation in these cells, a process that may contribute to understanding of the inflammatory host response to the vector (14, 16, 20).

A549 pulmonary epithelial cells express IL-8 after stimulation with tumor necrosis factor (TNF)-, phorbol myristate acetate (PMA), and IL-1 (20). A549 cells are a valid model for studing the events occurring in airway epithelial cells. CMV and respiratory syncytial virus (RSV) infections are known to induce production of IL-1 in mononuclear cells (21, 22), and RSV infection also induces IL-8 gene activation in epithelial cells (23, 24). We theorized that a potential mechanism for IL-8 induction by virus infection is through primary induction of IL-1, which then induces IL-8. If IL-1 does induce IL-8, the specific IL-1 receptor antagonist (IRAP) could therefore block IL-8 production. The blocking of IL-8 induction by IRAP has been reported by several investigators (25, 26). The present study was done to test the hypothesis that the induction of IL-8 in A549 cells by the adenovirus vector Av1LacZ4 is inhibited by preincubation with human IRAP.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Experimental Protocol

A549 pulmonary epithelial cells derived from a human alveolar cell carcinoma (American Type Culture Collection, Rockville, MD) were seeded into tissue-culture plates at a density of 5 × 10⁴ cells/cm² in Dulbecco's modified Eagle's medium (DMEM) (Gibco/BRL, Life Technologies Inc., Grand Island, NY) containing 10% lipopolysaccharide (LPS)-free fetal bovine serum (FBS) (Hyclone, Logan, UT), 100 U/ml penicillin, and 100 mg/ml streptomycin (Sigma Chemical Co., St. Louis, MO). The cells were grown to confluence at 37°C in humidified 95% air/5% CO₂. On the day of infection, the cell supernatants were discarded and fresh medium was added. The recombinant adenovirus vector was then added to the monolayer according to the infection protocol. Before the adenovirus vector was added, half of the plates were treated with 50 ng/ml of IRAP. Other A549 cells were stimulated with recombinant human IL-1 (hIL-1) (Endogen, Boston, MA) in concentrations of 0.2, 2, 20, and 200 pg/ml. For blocking experiments, cell cultures were pretreated with IRAP protein (a kind gift from Synergen, Boulder, CO) at concentrations of 10, 100, 1,000, and 10,000 pg/ml, which were applied 90 min before implementing the adenoviral infection or adding the cytokine stimulant. TNF- at a concentration of 100 U/ml (Endogen) was used as positive control. At 24, 48, and 96 h, cell-free supernatants were collected and cells were trypsinized and counted with a hemocytometer, and the cell viability was determined by trypan blue exclusion. Cells were also harvested for RNA extraction.

Virus Vector Construct

The Av1LacZ4 vector carries a nuclear-targeted LacZ gene. Av1LacZ4 is a first generation, E1a, E3-deleted recombinant adenovirus vector derived from Ad5. The heterologous minigene consists of the Rous sarcoma virus promoter driving expression of the nuclear targeted -galactosidase encoding sequence. The construction, purification, and titration of Av1LacZ4 have been described elsewhere (14). The vector was stored in virus dialysis buffer (10 mM Tris-HCl [pH 7.4], 1 mM MgCl₂, 10% glycerol) at 70°C until used.

Infection Protocol

Immediately before infection, Av1LacZ4 was thawed and added to the cells in fresh DMEM containing 2% FBS (Hyclone), 100 U/ml penicillin, and 100 mg/ml streptomycin. The vector concentration was calculated on the basis of a desired MOI of 50/cell and a total estimated cell density of 2.5 × 10⁵/cm² when the monolayer was confluent. This concentration was determined from previous cell-count studies (data not shown). After 90 min of exposure to the virus vector, the FBS concentration in the medium was corrected to 10%.

Enzyme-Linked Immunosorbent Assay

IL-8 and IL-1 were quantitated with a direct sandwich enzyme-linked immunosorbent assay (ELISA) method (20, 29). Each test sample was assayed in triplicate. A standard titration curve was obtained by making serial dilutions of human recombinant IL-8 protein (R&D Systems, Minneapolis, MN). Light absorption was measured at 450 nm on a THERMOmax microplate reader (Molecular Devices, Inc., Menlo Park, CA). The lower limit of sensitivity for this ELISA was 20 pg/ml.

Concentration of Supernatants

Concentration of supernatants from control cells, cells infected with the virus vector, and cells treated with TNF- was done with Centriprep concentrators (Amicon Inc., Beverly, MA). In essence, 15 ml of supernatant was concentrated in a Centriprep-10 through a three-step centrifugation procedure. The final-to-original concentration ratio was 15:1.

Membrane Preparation

Membrane preparations for control cells, cells infected with the virus vector, and cells treated with TNF- were isolated after 96 h of culture, through a modification of a detergent solubilization method described by Weissman (30).

Northern Blot Analysis

Total RNA was recovered by modification of the acid phenol extraction method of Chomczynski and Sacchi (20, 31). Lysis buffer (4 M guanidinium isothiocyanate, 25 mM sodium citrate [pH 7], 0.1 M -mercaptoethanol) was added to each well to extract RNA (0.5 ml/well). Purified RNA was stored at 80°C until further purification was done with Phase Lock Gell II (5 Prime  3 Prime, Inc., Boulder, CO) to optimize the recovery of RNA from the organic extraction according to the manufacturer's directions. RNA was quantified by measuring absorbance at 260 nm after dilution in 10 mM Tris, 0.1 mM ethylenediaminetetraacetic acid (EDTA) (pH 8.0). Aliquots of 15 mg of total RNA were denatured with 6 M glyoxal (40% ethanedial in aqueous solution) (Sigma), dimethylsulfoxide, and 0.1 M sodium phosphate (pH 7.0). The glyoxal solution was deionized (32). Samples containing equal amounts of RNA were fractionated on a 1.2% agarose gel and transferred to nylon membranes (Hybond, Amersham International, Amersham, UK). After crosslinking, the Hybond filters were stained with 1% methylene blue in 0.3 M sodium acetate to assess the integrity of the RNA and to verify the uniformity of loading.

The Hybond filters were then prehybridized for 2 h at 65°C in 5× saline-sodium phosphate-EDTA (SSPE) (1× SSPE = 150 mM NaCl, 10 mM NaH₂PO₄, and 1 mM EDTA [pH 7.4], 5× Denhardt's solution (1× Denhardt's solution is 0.02% bovine serum albumin, 0.02% Ficoll, 0.02% polyvinylpyrrolidone), 0.1% sodium dodecylsulfate (SDS), and 100 µg/ml salmon sperm DNA. Hybridization was performed overnight in a similar prehybridization solution with a [³²P]deoxycytosine triphosphate ([³²P]dCTP)- labeled 0.3-kb complementary DNA (cDNA) clone of human IL-8 mRNA (a kind gift from Dr. Ivan J. D. Lindley, Sandoz Forschungsinstitut, Austria), or a 0.4-kb 3' untranslated region (UTR) fragment of a human -actin cDNA clone.

The cDNAs were [³²P]dCTP-labeled (Amersham) with a random-primer DNA labeling system (T⁷ QuickPrime Kit; Pharmacia LKB Biotechnology, Piscataway, NJ), and the oligonucleotide (for Southern blot analysis) was labeled by 5' end labeling. After hybridization, blots were washed twice at room temperature with solutions of increasing stringency (standard saline citrate with 0.1% SDS) and at increasing temperature until an adequate background reading was obtained. All filters were wrapped in plastic wrap and exposed to a phosphorimaging storage screen for 20 min before peak volume quantitation, which was done with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA), using ImageQuant software (33). Additionally, autoradiograms for IL-8 and -actin were generated by exposure of the filters to Kodak XAR-2 film (Eastman Kodak, Rochester, NY) at 80°C overnight.

Duplicate Hybond filters were hybridized with ³²P-labeled IL-8 cDNA. After quantification with the PhosphorImager and autoradiography, the same filters were stripped and rehybridized with ³²P-labeled human -actin cDNA. In order to normalize for loading quantities, the data were expressed as:

(1)

which represents the ratio of IL-8-mRNA from treated cells (txt) to IL-8-mRNA from control cells (Co), divided by the ratio of -actin units of mRNA from treated cells to -actin units of mRNA from control cells.

Reverse Transcription-Polymerase Chain Reaction

Reverse transcription (RT) was done with Moloney murine leukemia virus reverse transcriptase, after which the polymerase chain reaction (PCR) was performed (34), with both procedures done according to the manufacturer's directions (Gibco/BRL, Life Technologies, Inc., Grand Island, NY), using a Perkin-Elmer (Norwalk, CT) PCR apparatus. In brief, the RT sample had a volume of 19.5 µl and the reaction was performed for 1 h at 37°C. Two microliters of reverse-transcribed product were used for the PCR (30 cycles) reaction of the pro-IL-1 sequence. From this PCR product, 2 µl were used for a nested PCR reaction of the secreted IL-1 sequence. A standardization protocol was implemented with -actin PCR as internal control. The primer used for the pro-IL-1 upstream primer was 5'-GCCCTAAACAGATGAAGTGTCC-3', and that for the downstream primer was 5'-ATTGCATGGTGAAGTCAGTTATATC-3'. For the nested PCR of the secreted sequence of IL-1, a 5'-GCTGATGGCCCTAAACAG-3' upstream primer and 5'-GAAGACGGGCATGTTTTC-3' downstream primer were used. Primers for IL-1 were 5'-GTAAGCTATGGCCCACTC-3' and 5'-GAAATAGTTCTTAGTGCCGTG-3' (sense and antisense, respectively). RT-PCR analysis for pro-IL-1 with subsequent nested PCR for the secreted sequence of IL-1 was done with total RNA from uninfected A549 (control) and infected cells at 96 h of culture, together with U937 cells untreated and treated with PMA (10 ng/ml) for 24 h. An IL-1 cDNA was used as a positive control for the PCR.

Southern Blot Analysis

Agarose gel (1%) electrophoresis was done with 20 µl of the PCR product. An aliquot of DNA mass ladder (4 µl; Gibco/BRL, Life Technologies Inc., Gaithersburg, MD) was added for the recognition of bands.

The agarose gel was denatured by soaking it for 20 min in 0.4 N NaOH, 1 M NaCl, transferring the bands to nylon membranes (Hybond), and cross-linking. The Hybond membranes were prehybridized and hybridized according to the same used for the Northern blot analysis, with an IL-1 oligonucleotide.

LacZ Staining of Cytospin Preparations

Cells were removed from the culture plates by trypsin digestion, washed with phosphate-buffered saline (PBS), and resuspended at a concentration of 25 × 10⁵ cells/200 µl. Cytospin preparations were made on a cytocentrifuge (Shandon Southern, Sewickley, PA) at 700 rpm for 5 min, air dried for 1 h, and stored at 70°C until used. Slides were fixed with 0.5% glutaraldehyde for 10 min at room temperature, washed with PBS, and stained for 3 h with 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal) stain (5 mM potassium ferrous cyanide, 5 mM potassium ferric cyanide, 2 mM magnesium chloride, 1 mg/ml X-Gal, 0.02% NP40 in PBS). Av1LacZ4 expresses a nuclear-targeted LacZ gene. Therefore, A549 cells transduced by Av1LacZ4 have blue-staining nuclei after histochemical staining for -galactosidase.

Statistical Analysis

Statistical analysis, with replicate experiments as covariants (n = 3 experiments with three replicates), was done at each time point by analysis of variance, and subsequent analysis of mean pairs was done with Student's t test. Data are presented as means ± SD, and results were considered significant at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

IL-8 Gene Expression in Infected Cells

Av1LacZ4 infection of the A549 epithelial monolayer at an MOI of 50 produced increased IL-8 gene expression. IL-8 gene expression, as measured from mRNA levels and compared with that in uninfected cells, was increased by 7.1 ± 6.2-fold at 24 h (P = 0.11), by 6.1 ± 4.4-fold at 48 h (P = 0.2), and by 9.3 ± 5.3-fold at 96 h (P < 0.06) (Figure 1). Stimulation of A549 cells with TNF- at a concentration of 100 U/ml increased the IL-8 mRNA level by 24 h to 6.2-fold that of unstimulated cells, after which the levels decreased to near baseline by 96 h (data not shown).

View larger version (33K): [in this window] [in a new window]   Figure 1.   Northern blot analysis showing downregulation of IL-8 gene expression in infected A549 cells by pretreatment with 50 ng/ml IRAP. Cells were cultured as described in MATERIALS AND METHODS and were harvested at 24, 48, and 96 h after collection of supernatants. Human IL-8 cDNA hybridization blots are seen in upper row of autoradiograms, and human -actin cDNA hybridization blots are seen in lower row of autoradiograms. Bar graphs depict peak volume quantitation ratio of cytokine mRNA from treated cells to cytokine mRNA from control cells, divided by the ratio of -actin mRNA from treated cells to -actin mRNA from control cells (for normalization of loading) from three experiments (n = 3), with three replicates each. Peak volume quantitation was done with a PhosphorImager, using ImageQuant software. A significant statistical difference (asterisk) is seen at 96 h.

Demonstration of Extracellular IL-8

A549 cells constitutively produced low levels of IL-8. Infection of A549 cells with virus vector resulted in a significant increase in IL-8 secretion above baseline levels in a time-dependent fashion (Figure 2). At 24 h, the IL-8 level in the virus-infected group was 0.47 ± 0.08 ng/ml, whereas the level in the uninfected group was 0.17 ± 0.05 ng/ml (P < 0.001). At 48 h, the IL-8 level in the virus-infected group was 0.96 ± 0.18 ng/ml, whereas the level in the uninfected group was 0.46 ± 0.12 (P < 0.001), and by 96 h the levels in the two groups were 4.8 ± 0.22 ng/ml and 1.4 ± 0.09 ng/ml, respectively (P < 0.001) (Figure 2). The TNF-- stimulated cells had a significantly greater concentration of IL-8 at all time points (7.1 ± 0.2 ng/ml at 24 h; 6.3 ± 0.3 ng/ml at 48 h; 7.9 ± 0.1 ng/ml at 96 h).

View larger version (27K): [in this window] [in a new window]   Figure 2.   ELISA results showing downregulation of IL-8 levels after infection by virus vector in A549 cells pretreated with 50 ng/ml IRAP. A significant statistical difference in vector-induced IL-8 levels as compared with the levels in control cells is seen at 24, 48, and 96 h. Significant downregulation of IL-8 production by pretreatment with IRAP is seen at 48 and 96 h (asterisks) in the vector infected cells as compared with the uninfected cells. Cells were cultured as described in MATERIALS AND METHODS and collection of supernatants was done at 24, 48, and 96 h. Bar graphs depict the average of IL-8 levels from three experiments (n = 3) with three replicates each.

Pretreatment of Virus Vector-Infected A549 Airway Epithelial Cells with IRAP

IRAP downregulated by one third the IL-8 gene activation induced by the virus vector, with the decrease reaching significance at 96 h (Figure 1). Protein levels of IL-8 were also significantly downregulated at both 48 and 96 h (P < 0.05) (Figure 2). In one experiment, IL-8 protein levels were standardized by cell counting, which showed consistency. IRAP did not affect IL-8 mRNA induced by TNF- (data not shown), and IRAP alone did not have any affect on the constitutive production of IL-8.

Blocking Effect of IRAP on IL-8 Gene Activation in IL-1-Stimulated Respiratory Cells

IL-1 could not be detected with ELISA in the supernatant from infected cells. To demonstrate that IL-1 concentrations below the lower limit of the ELISA were able to induce mRNA and secretion of IL-8, we stimulated A549 cells with IL-1 at concentrations of from 0.2 to 20 pg/ml. There was a dose-related increase in IL-8 expression with increasing concentrations of IL-1 (Figure 3). A dose of IL-1 as low as 0.2 pg/ml was able to induce a statistically significant increase in IL-8 concentration as compared with the concentration for unstimulated cells. Pretreatment of IL-1-stimulated cells with IRAP reduced IL-8 production in a dose-related manner (Figure 3).

View larger version (41K): [in this window] [in a new window]   Figure 3.   Downregulation of IL-8 by pretreatment with IRAP in IL-1-stimulated A549 cells. A significant dose-related increase is seen in IL-8 production with increasing concentrations of IL-1 from 0.2 to 20 pg/ml (double asterisks; P < 0.02). This increase was downregulated by pretreatment with IRAP at a 50-fold excess over IL-1 concentrations (asterisk; P < 0.03). Cells were cultured as described in MATERIALS AND METHODS, and collection of supernatants was done at 24 h. Bar graphs depict the average of IL-8 levels from two experiments (n = 2) with three replicates each.

RT-PCR and Southern Blot Hybridization for IL-1 from Infected Cells

The expected 713-bp size band for pro-IL-1 was identified, and a 404-bp band was identified for the IL-1 cDNA after the nested PCR (Figure 4). We found increased (and sustained) expression of IL-1 mRNA in the infected cells as compared with the untreated uninfected cells. Recognition of bands for IL-1 (pro- and cDNA) was done through Southern blot analysis with a 300-bp cDNA for IL-1 (Figure 5). No bands were seen after RT-PCR analysis done with primers for IL-1.

View larger version (33K): [in this window] [in a new window]   Figure 4.   RT-PCR gel showing a 713-bp DNA band representing the pro-IL-1 band for infected A549 cells as well as U937 cells untreated and treated with PMA for 24 h (as positive control). There was no band for uninfected (control) A549 cells. Agarose gel electrophoresis with ethidium bromide staining shows RT-PCR results with pro-IL-1 primers, as described in MATERIALS AND METHODS (asterisk: DNA mass ladder recognizing fragments of 2,000, 1,200, 800, 400, and 200 bp).

View larger version (34K): [in this window] [in a new window]   Figure 5.   Autoradiograms of Southern blots of nested PCR products, showing homologous recognition of pro-IL-1 (713 bp) and secreted IL-1 (300 bp) bands. Bands are seen for virus vector-infected A549 cells and U937 cells, untreated and treated with PMA. No band is seen for uninfected (control) A549 cells. Hybridization of the nested PCR products was done on nylon membranes, using an IL-1  oligonucleotide, as described in MATERIALS AND METHODS.

Determination of IL-1 Production by ELISA of Concentrated Supernatants and Membrane Preparations

We were unable to detect IL-1 in unconcentrated supernatants. The concentrated supernatants from vector-infected cells showed significantly higher levels of IL-1 at 96 h (37 ± 1 pg/ml) than did concentrated supernatants from uninfected control cells, in which IL-1 was undetectable.

Av1LacZ4 Transduction Rate

The rate of transduction of A549 cells with Av1LacZ4 at 24, 48, and 96 h after exposure to the virus was 47.5 ± 15.6%, 80 ± 5.09%, and 85.5 ± 3.93%, respectively, indicating efficient transduction by the virus.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study support the hypothesis that the adenovirus vector Av1LacZ4 partly induces IL-8 transcription and secretion in A549 cells through IL-1 induction. By blocking the IL-1 receptor with IRAP and partly inhibiting the adenovirus vector-induced expression and release of IL-8, we showed the relationship of IL-1 to IL-8 induction. Other researchers have showed a mechanism of IL-8 production through primary IL-1 induction. Lukacs and colleagues (34) demonstrated significantly decreased levels of IL-8 on Day 5 of culture after IRAP pretreatment at 25 ng/ml. DeForge and coworkers (26) addressed the question of whether IRAP would suppress IL-8 produced by LPS-stimulated human blood monocytes, and found downregulation of IL-8 by IRAP. Kaplanski and associates (27) verified a similar mechanism of IL-8 production through IL-1 in activated platelets, and Porat and coworkers (34) showed that Borrelia burgdorferi induces IL-8 by IL-1 induction, which can be regulated by IRAP.

We were able to demonstrate expression and secretion of IL-1 by virus-infected cells through RT-PCR, and to measure secreted IL-1 protein in concentrated supernatants from treated cells, although we failed to detect IL-1 in the supernatants of infected cell with the usual ELISA technique. Smith and colleagues (35) have also reported IL-1 production by A549 cells. We showed that a concentration below 15 pg/ml of IL-1 in the supernatant was sufficient to induce IL-8 production, and that the resulting IL-8 levels could be reduced by pretreating the cells with IRAP. Furthermore, we found that the adenovirus vector used in our study induced sustained expression of IL-1 and IL-8 for up to 96 h, which contrasts markedly with the peak of cytokine activity occurring at 24-48 h when A549 cells were stimulated with IL-1 or TNF-. RT-PCR analysis for IL-1 mRNA and ELISA measurements of concentrated supernatants for secreted IL-1 polypeptide showed that this cytokine was present for up to 96 h after infection of A549 cells, whereas no mRNA or antigenic protein was found in untreated cells. The viral effect on IL-1 induction in cells infected in vitro is consistent with the finding in previous studies that showed a direct stimulation of IL-1 and IL-1 production by other viruses, such as CMV and RSV (21, 36, 37).

The immune response to wild type adenovirus is modulated by several viral gene products (13). It is not clear how the virus stimulates the induction of IL-1. IL-1, TNF, and PMA induce IL-8 by inducing a nuclear factor (NF)- B-like factor that complexes the region between 80 and 71 bp of the IL-8 promoter (38). Another factor, binding to the region between 94 and 81 bp, and which was proposed (39) as a pathway for cooperative IL-1 gene expression by IL-1, TNF, or PMA, is an NF-IL-6-like factor coupled to CCAAT/enhancer binding protein (C/EBP). The IL-6 gene is inducible by IL-1 or TNF (13). IL-1 induces translocation of the NF-B-like factor, replacing the weaker C/EBP-like factor and leading to expression of IL-8.

Our current study provides an important contribution to the understanding of viral induction of IL-8 by demonstrating IL-1 activation (which activates translocation of the NF-b-like factor [39] and leads to IL-8 gene expression). IRAP could be effective in the treatment of patients with viral infections when IL-8 induction is the main pathway inducing inflammation. Specifically, IRAP can be given prophylactically to patients undergoing gene therapy in order to prevent the activation of IL-8 as well as other effects of IL-1. McCoy and associates (40) were unable to show blockage of inflammation induced by the virus vector AdRSV, which contained a human cDNA IRAP. We performed a 90-min preincubation of A549 cells with IRAP before infection or IL-1 stimulation. Moreover, we used IRAP in a 50-fold excess over IL-1 concentrations, owing to the potent inflammatory effect of low concentrations of IL-1 (0.2 pg/ml). The peak levels of expression of IRAP in McCoy and colleagues' study were 5.5 ng/ml in the BALF of male CBA/J mice, but they did not measure the level of IL-1 in the fluid. To be able to block the proinflammatory effects of IL-1, IRAP has to be present in a sufficient excess (41, 42). It can be hypothesized that if McCoy and colleagues' mice had been pretreated with IRAP or actually inoculated with virus having a promoter that amplifies the transcription of human IRAP in order to induce enough copies of the IRAP protein, it would have been present in a sufficient excess to block IL-1.

Other models of induction of IL-8 could be studied by using IRAP to block IL-8 production and demonstrate that such production occurs through IL-1 activation. Such models would be useful for studying mechanisms of inflammation in different diseases in which IL-8 is expressed.

In summary, our study contributes to the understanding and modulation of virus-induced inflammation by suggesting a role for IL-1 in early induction of the cytokine network that leads to airway inflammation after the administration of adenoviral vectors.