ABSTRACT
Infection with adenovirus serotype 7 (Ad7) frequently causes lower respiratory pneumonia and is associated with severe lung inflammation and neutrophil infiltration. Earlier studies indicated release of proinflammatory cytokines, specifically interleukin-8 (IL-8), by pulmonary epithelial cells following infection by Ad7. However, the mechanism of IL-8 induction by Ad7 is unclear. We have explored the role of the Ras/Raf/MEK/Erk pathway in the Ad7-associated induction of IL-8 using a model system of A549 epithelial cells. We found that Ad7 infection induced a rapid activation of epithelial cell-derived Erk. The MEK-specific inhibitors PD98059 and U0126 blocked Erk activation and release of IL-8 following infection with Ad7. Treatment with PD98059 is cytostatic and not cytotoxic, as treated cells regain the ability to phosphorylate Erk and secrete IL-8 after removal of the drug. The expression of a mutated form of Ras in A549 epithelial cells blocked the induction of IL-8 promoter activity, and MEK inhibitor blocked induction of IL-8 mRNA. These results suggest that the Ras/Raf/MEK/Erk pathway is necessary for the Ad7 induction of IL-8 and that induction occurs at the level of transcription. Further, the kinetics of Erk activation and IL-8 induction suggest that an early viral event, such as receptor binding, may be responsible for the observed inflammatory response.
Adenovirus (Ad) infections cause pneumonia and disseminated disease in both immunocompromised and nonimmunocompromised hosts. There are at least 49 types of Ad, as defined by serological methods. While many of the Ad serotypes cause infectious syndromes, most are mild upper respiratory infections. In contrast, Ad type 7 (Ad7) has been implicated as a cause of a more severe lower respiratory tract infection. Ad7 has been isolated during episodes of pneumonia in military recruits and children (15). Diffuse alveolar damage similar to that seen in acute respiratory distress syndrome is often a consequence of Ad7 infection and can lead to permanent damage to lung tissue as a result of the massive inflammation (39).
Ad-based vectors have been employed for in vitro and in vivo delivery of genes into mammalian cells. Most vectors used in gene therapy are derived from the Ad5 serotype. However, the use of Ad5-based vectors in nonhuman primates was associated with a profound inflammatory response (41). While it may be a potential delivery vector, Ad7 is not currently being used as a human gene therapy vector. Ad7 employs a unique but unknown cellular receptor for viral fiber protein. However, this feature may make Ad7 a potential route to alleviate the limited tissue tropism of existing Ad vectors. Thus, a better understanding of Ad-triggered inflammation is important to alleviate the symptoms of wild-type Ad infection and to reduce the potential side effects of Ad gene therapy vector administration.
Alveolar epithelial cells, which are infected during acute Ad infection, are likely mediators of the inflammatory response to Ad (1). The CXC chemokine interleukin-8 (IL-8) is an important mediator of the inflammatory response to many stimuli, including viruses (5). Additionally, IL-8 may be important in the pathophysiology of asthma and obstructive lung disease, since IL-8 levels are increased in patients with these chronic inflammatory disorders (35). IL-8 has also been measured in bronchoalveolar lavage fluid from patients with acute respiratory distress syndrome (25) and secretions from patients with Ad19-associated keratitis (12). Most importantly, IL-8 is also a major neutrophil chemoattractant and activator (reviewed in reference 5). Neutrophils are prominent in lavage fluid in many animal models of Ad infection, as well as upon exposure of humans and nonhuman primates to genetically modified Ad5 vectors known to induce IL-8 in vitro (6, 10, 20, 21, 41).
Our earlier studies showed that human pulmonary epithelial-like cells release IL-8 upon infection with Ad7 (7); however, the mechanism of IL-8 induction is not understood. Other studies of infectious processes caused by agents including respiratory syncytial virus, lipopolysaccharide, mycobacterium, and Pseudomonaspyocyanin have indicated a role for the cellular kinase extracellular regulated kinase (Erk) in the induction of IL-8 by host cells (11, 13, 31). Erk activation is initiated by a wide variety of growth factor receptors through a specific set of signal transduction events. These events culminate in the receptor recruitment of a guanine nucleotide exchange factor into proximity with plasma membrane-localized Ras to catalyze Ras activation by GTP binding (reviewed in reference 38). Once activated, Ras-GTP initiates a cascade of serine-threonine kinases, beginning with Raf, which phosphorylates and activates mitogen-activated protein kinase (MAPK)-activating kinase (MEK). MEK phosphorylates and activates Erk, which translocates to the nucleus to activate transcription factors, such as NF-κB and AP-1. Both of these transcription factors are important for the induction of IL-8 transcription (reviewed in reference 5). Induction of IL-8 does not exclusively involve signaling through activation of the Erk pathway. For example, PD98059 (the inhibitor of Erk used in our study) does not block tumor necrosis factor (TNF)- and IL-1-stimulated IL-8 production from pulmonary vascular endothelial cells (23) or TNF-stimulated IL-8 production from neutrophils (42). Likewise, PD98059 had a minimal effect on group B streptoccoccal induction of IL-8 in these cells (2).
Earlier experiments using an epithelial cell model and infection with an Ad5-based gene therapy vector demonstrated activation of the Erk pathway soon after infection and of IL-8 production at later time points. However, the relationship between Erk induction and IL-8 production is not understood (10). Our laboratory has previously demonstrated that the production of IL-8 by host cells following Ad7 infection occurs more rapidly than that induced by Ad5-based gene therapy vectors, i.e., within 2 rather than 24 h (7). We hypothesized that the difference in the kinetics of IL-8 production might be due to differences in signal transduction pathways initiated by Ad7. Therefore, we sought to address the role of the Erk pathway in Ad7-associated induction of IL-8.
In this report, we demonstrate that activation of the Ras/Raf/MEK/Erk pathway correlates with induction of IL-8 by Ad7. Using two specific inhibitors of the Erk pathway, we show that Erk activation is essential for IL-8 production stimulated by Ad7. Inhibition of the Erk pathway and IL-8 production by these inhibitors can be reversed and therefore is not due to a nonspecific cytotoxic effect. Additionally, we observed that cellular expression of a dominant-negative Ras molecule blocks the ability of the cells to induce IL-8 promoter activity upon Ad7 infection and that chemical inhibition of Erk blocks IL-8 mRNA induction by virus. We conclude that Ad7-mediated activation of the Erk pathway is causally related to the subsequent production of IL-8, which can contribute to inflammatory disease in the host.
MATERIALS AND METHODS
Cells and viruses.The human pulmonary epithelial cell line A549 was used for these studies. This cell line, derived from a patient with lung carcinoma, has characteristics of alveolar type II cells and was obtained from the American Type Culture Collection, Manassas, Va. (ATCC CCL-185) (30). The cells were maintained in Dulbecco's modified Eagle's medium (DMEM) with 2 mMl-glutamine and 80 μg of gentamicin/ml plus 10% fetal bovine serum (FBS) at 37°C in a humidified incubator at 5% CO2. Ad7 was obtained from the American Type Culture Collection (ATCC VR-7). Virus was propagated as described previously (7). Infected A549 cells were incubated for 3 to 4 days in DMEM plus 10% FBS; virus was harvested by three freeze-thaw cycles and purified by ultracentrifugation in CsCl gradients as described previously (24). The purified virus preparation was dialyzed extensively against phosphate-buffered saline (PBS), pH 7.2, plus 10% glycerol, titered by a cytotoxic plaque assay (27), and stored at −80°C.
Kinase assays, immunoprecipitation, SDS-PAGE, and immunoblotting.A549 cells grown on 10-cm-diameter dishes for 18 h in serum-free DMEM to approximately 90% confluency were infected with Ad7 at a multiplicity of infection (MOI) of 50 in 3 ml of DMEM minus FBS and incubated at 37°C and 5% CO2 for various times. Phorbol 12-myristate 13-acetate (PMA; Sigma) at 100 ng/ml (unless otherwise stated) was used as a positive control, and mock-infected negative control cells were exposed to equivalent volumes of virus-free buffer (PBS plus 10% glycerol). When appropriate, the mock-infected cells were also treated with inhibitor solvent (dimethyl sulfoxide [DMSO]). Cells treated with various concentrations of the specific MEK inhibitor PD98059 (Calbiochem) (3, 16) were incubated for 30 min at 37°C and 5% CO2 in 1 ml of serum-free DMEM prior to the addition of Ad7 or PMA. Where indicated, PD98059 was washed out of the cells by three successive washes in PBS. Negative control cells were preincubated with equivalent amounts of DMSO. After infection, A549 cells were washed three times in ice-cold PBS and lysed in cold lysis buffer (150 mM NaCl; 50 mM Tris [pH 8.0] 10 mM [each] EDTA, NaF, and NaP∼P; 1% NP-40; 0.5% Na deoxycholate; 0.1% sodium dodecyl sulfate [SDS]; 3 mM sodium vanadate; 10 μg of aprotinin and leupeptin/ml). The cells were scraped from the tissue culture dishes, and the lysates were transferred to Eppendorf tubes. Postnuclear lysates were immunoprecipitated overnight at 4°C with 3 μg of rabbit anti-Erk1/2 antibody (Upstate Biotechnology, Inc., Lake Placid, N.Y.), followed by incubation with 25 μl of Sepharose-bound fusion protein A/G (Pierce Inc., Rockford, Ill.) for 1 h at 4°C. Control lysates were comparably immunoprecipitated with an equivalent amount of normal rabbit immunoglobulin (NRIg) after PMA stimulation. The immunoprecipitated material was washed five times with lysis buffer containing 1 mM sodium vanadate, resuspended in 50 μl of SDS sample buffer (60 mM Tris [pH 6.8], 10% glycerol, 2.3% SDS), and heated to 95°C for 5 min. The samples were separated by SDS–12.5% polyacrylamide gel electrophoresis (PAGE) and electrophoretically transferred to nitrocellulose membranes. To detect activated phosphorylated Erk (22, 36), the filters were immunoblotted with a mouse monoclonal antibody specific for both the p44 and p42 diphosphorylated forms of Erk (Upstate Biotechnology, Inc.). The filters were developed with horseradish peroxidase-labeled goat anti-mouse IgG (Kirkegaard and Perry Laboratories). The same filter was stripped with 2% SDS in 50 mM Tris, pH 6.0, to remove the primary antibody and subsequently reprobed with polyclonal anti-Erk1/2 antibody (Upstate Biotechnology, Inc.) that recognizes both phosphorylated and nonphosphorylated Erk. Blots were developed and quantified using the LumiImager F1 system and LumiAnalyst software (Boehringer Mannheim GmbH, Mannheim, Germany).
In vitro kinase assay.The protocol to immunoprecipitate Erk was identical to that described above. Immunoprecipitated Erk was resuspended in 10 μl of kinase assay dilution buffer (25 mM HEPES [pH 7.0], 0.15 M NaCl, 10 mM MnCl2, 1 mM NaVO4, 5 mM dithiothreitol, 0.5% NP-40). A reaction buffer composed of 75 mM MgCl2, 500 μM ATP, 10 mCi [32γ-P]ATP, and 2 mg of myelin basic protein (MBP)/ml was added to the bead-bound Erk (20 μl/reaction) and incubated for 15 min at 30°C. The reaction was stopped by adding 5× SDS sample buffer and incubating the mixture at 95°C for 5 min. The material in the kinase reaction was separated by SDS–15% PAGE and transferred to a nitrocellulose membrane before being sealed and exposed to a PhosphorImager screen. The bands representing phosphorylated MBP were analyzed using Imagequant software after being scanned on a Storm PhosphorImager (Molecular Dynamics).
IL-8 ELISA.After overnight growth at 37°C and 5% CO2 on 12-well plates in DMEM plus 10% FBS, 8.5 × 105 A549 cells/well were infected with Ad7 at an MOI of 50 in 0.3 ml of DMEM plus 0.5% FBS and incubated for 1 h at 37°C and 5% CO2. Subsequently, 0.7 ml of DMEM plus 10% FBS was added to each well, and the plates were returned to the incubator. PMA (100 ng/ml) served as a positive control, PBS plus 10% glycerol was used as a negative control, and all controls were treated similarly to Ad-infected cells. Cells exposed to various concentrations of PD98059 or U0126 (Calbiochem) were incubated for 1 h in DMEM plus 0.5% FBS prior to the addition of Ad7 or PMA. The final concentration of PD98059 or U0126 was maintained throughout the experiment. For specified experiments, the PD98059 was washed out by three successive washes in PBS. Negative control cells were treated with DMSO. At various times, the culture supernatants were collected and analyzed for IL-8 protein by a sandwich ELISA (R&D Systems, Minneapolis, Minn.).
Transfections.A549 cells were seeded in 25-cm2flasks at 105 cells per flask and grown overnight. The cells were transfected with 1 μg of the −162 to +44 IL-8 lucerifase reporter plasmid containing the IL-8 5′ promoter region linked to a luciferase reporter gene (a gift of Allen Brasier, University of Texas Medical Branch, Galveston) (19). The transfections were carried out with GenePorter 2 reagent (Gene Therapy Systems, San Diego, Calif.). Where indicated the cells were cotransfected with 5 μg of inactivated Ras (N17Ras [26]). After a 24-h incubation period, the transfected cells were washed to remove the transfection reagent, and fresh DMEM plus 10% FBS was added. After an additional 24-h incubation, the medium was replaced with serum-free DMEM. After 16 h of serum starvation, the transfected cells were treated with the MEK inhibitor U0126 (25 μM) or DMSO and stimulated with Ad7 (MOI = 50) or PMA. The cells were incubated for 6 h before lysis. Postnuclear lysates were assessed for induced luciferase activity with an assay kit from Promega (Madison, Wis.).
Isolation of RNA and Northern blot analysis.Cells were infected as described above prior to harvest in the presence or absence of 25 μM PD98059. All solutions used for RNA preparation were treated with diethyl pyrocarbonate at 1:1,000. The cells were scraped off the plates, pelleted by centrifugation, and washed with PBS. Whole cellular RNA was prepared using Trizol reagent (Gibco BRL, Grand Island, N.Y.). The RNA was measured by UV absorbance at 260 nm and stored at −70°C prior to Northern analysis. RNA was dissolved in a MOPS (morpholinepropanesulfonic acid)-formaldehyde-formamide solution and heated for 5 min at 60°C. Samples (5 μg/lane) were run on an agarose-formaldehyde (1.5% agarose–2.2 M formaldehyde–(MOPS) gel. The gel was stained with ethidium bromide.
Following overnight capillary transfer to a Gene Screen (Dupont, NEN, Boston, Mass.) membrane, the RNA was probed with a 32P end-labeled 30-bp oligomer complementary to the IL-8 protein coding region (7). Fresh buffer was added to the blot along with the amount of probe required to give 106 counts/ml. Salmon sperm DNA was added to block nonspecific binding. After incubation, the blot was removed, washed, and exposed, and signals were quantitated using a Storm PhosporImager.
RESULTS
Induction of IL-8 protein by Ad7.We first investigated the ability of A549 human lung epithelial cells to respond to Ad7 infection by the production of IL-8 protein. Confluent A549 cells were either mock infected with viral diluent (PBS plus 10% glycerol), infected with Ad7 (MOI = 50), or exposed to PMA (100 ng/ml). Following various incubation times from 2 to 24 h, the supernatants were harvested and assessed for IL-8 levels using a sandwich ELISA. Ad7 infection of A549 cells caused a five- to eightfold increase in IL-8 production above that of mock-infected cells (Fig.1). Increased IL-8 was detectable within 2 h after infection with Ad7, reached maximal levels at 8 h, and persisted but did not increase through 24 h. The amounts of IL-8 induced by Ad7 were 28 to 45% of the amounts induced by 100 ng of PMA/ml and are similar to those seen in cells stimulated with TNF-α (7). These results confirm our previous model (7) and establish that Ad7 induces the release of proinflammatory IL-8 from A549 cells. The fold increase in IL-8 induction appears similar to that reported for respiratory syncytial virus-infected A549 cells (4).
Induction of IL-8 protein by Ad7. Confluent A549 cells were exposed to wild-type Ad7 at an MOI of 50 or to virus-free buffer (mock infection) for various times prior to measurement of supernatant IL-8 by ELISA. PMA (100-ng/ml)-exposed cells were used as a positive control. All measurements were performed in triplicate and are expressed as the mean + standard error of the mean of three separate experiments (see Materials and Methods).
Infection with Ad7 rapidly induces Erk phosphorylation.To determine if Ad7-associated production of IL-8 correlated with Erk activation, we first assessed Erk phosphorylation in confluent, serum-starved A549 cells infected with Ad7 at an MOI of 50. At various time intervals after infection, lysates of A549 cells were prepared and total Erk1/2 was immunoprecipitated as described in Materials and Methods. PMA stimulation served as a positive control for Erk phosphorylation and was allowed to continue for a full 60 min. Mock-infected lysates were prepared from cells at time zero and at 60 min. The immunoprecipitated Erk was analyzed by SDS-PAGE and by immunoblotting with an antibody recognizing the dually phosphorylated form of Erk (22, 36). The results (Fig. 2A and B) revealed that Erk phosphorylation and activation occurred within 5 min of Ad7 contact with the epithelial cells. Erk phosphorylation peaked at 10 min of exposure and returned to background by 60 min. We quantitated Erk phosphorylation by determining the ratio of phosphorylated Erk (Fig. 2A) to total Erk (Fig. 2B), and the results are shown in Fig. 2C. The ratio corrects for variations in the completeness of Erk immunoprecipitation. These results indicate that Ad7 infection elicits Erk activation, which precedes IL-8 production in A549 epithelial cells.
Kinetics of phosphorylation of Erk by Ad7. Serum-starved A549 cells were exposed to Ad7 at an MOI of 50 for various times prior to preparation of cellular lysates and Erk1/2 immunoprecipitation. PMA-exposed cells (60 min), mock-infected cells (either time zero or 60 min), and cells exposed to PMA but immunoprecipitated with NRIg (PMA-NRIg) were used as positive and negative controls. (A) Western blot performed with antibody specific for dually phosphorylated Erk1/2. (B) Same blot stripped and reprobed with pan-anti-Erk1/2 antibody. (C) Activation of Erk as determined by ratio of phospho-Erk to total Erk for each immunoprecipitate. In all cases, bands for both p44 and p42 Erks were included in the calculations. The data shown are representative of three separate experiments.
Inhibition of Ad induction of Erk by a chemical MEK inhibitor.In mammalian cells, Erk is activated by a sequential pathway initiated by GTP loading of Ras and is sustained by two additional kinases, Raf and MEK (38). PD98059 was developed as a specific MEK inhibitor (3, 16), which we utilized to determine the precise molecular relationship between Ad7-triggered Erk activation and IL-8 production in the epithelial cell line. To determine if activation of Erk by Ad7 occurred through the expected Raf/MEK pathway, we treated A549 cells with increasing doses of PD98059 prior to a 20-min infection with Ad7 or stimulation with PMA. Mock-infected cells were exposed to an equal volume of virus-free buffer containing the same concentration of inhibitor solvent (DMSO) as that used at the highest dose of inhibitor. Figure 3A shows the phosphorylation of Erk following stimulation or Ad7 infection in the presence of increasing dosages of PD98059. Figure 3B shows the same membrane reprobed for total Erk protein, and Fig. 3C shows the ratio of the two. Ad7-inducible Erk phosphorylation was reduced in a dose-dependent fashion by PD98059, and 25 μM PD98059 reduced Erk phosphorylation in Ad7-exposed cells to near-background levels. These amounts of PD98059 are consistent with earlier studies using the A549 cell line as a model to study Erk activation by epidermal growth factor (9).
Inhibition of Ad induction of Erk by MEK inhibitor. Serum-starved A549 cells preincubated with various doses of the MEK inhibitor PD98059 were exposed to Ad7 at an MOI of 50. PMA-stimulated or mock-infected cells preincubated with or without PD98059 were used as positive and negative controls. Cells exposed to PMA immunoprecipitated with NRIg (PMA-NRIg) were used as a control for immunoprecipitation. (A) Immunoblot of mock-infected, Ad7-infected, or PMA-stimulated cell extracts probed with anti-phospho-Erk1/2 antibody. (B) Same blot as in panel A stripped and reprobed with pan-anti-Erk1/2 antibody. (C) Ratio of pErk to Erk, determined as for Fig. 2. The data are representative of four separate experiments.
In our assay system, phosphorylation of Erk is used as an indicator of activated Erk, capable of eliciting transcription factor activation and hence IL-8 production. In order to confirm that Erk was indeed enzymatically activated during Ad7 infections, we performed in vitro kinase assays on immunoprecitipated Erk. As before, we prepared lysates of A549 cells either mock infected, Ad7 infected, or stimulated with PMA. The proteins within the in vitro kinase reactions were separated by SDS-PAGE, and the amount of 32P present in the MBP in vitro kinase substrate was quantitated by Storm image analysis. The results (Fig. 4) showed that mock-infected cells displayed a low level of activated Erk, and this activity was further reduced by treatment with PD98059. Ad7 infection of A549 cells resulted in an ∼20-fold increase in Erk enzymatic activity above that induced by mock infection. The induction by Ad7 of Erk activity was reduced to near-background level in the presence of PD98059. PMA stimulation elicited increased Erk activity in A549 cells, and this stimulation was likewise susceptible to inhibition by PD98059 treatment. These results indicate that analysis of phosphorylated Erk by immunoblotting with a phosphospecific antibody is equivalent to analysis of Erk enzymatic activity by in vitro kinase measurements. Additionally, these findings further establish that PD98059 treatment blocks Erk induction in A549 cells by PMA or Ad7 infection.
Ad7 induces Erk activity. A549 cells were mock infected, PMA stimulated, or Ad7 infected (MOI = 50) in the presence or absence of 25 μM PD98059 for 20 min prior to the preparation of lysates. The lysates were immunoprecipitated overnight with anti-Erk1/2 antibody or with rabbit IgG NRIg. Immunoprecipitated Erk1/2 was then analyzed for activity by an in vitro kinase assay as described in Materials and Methods. (A) Phosphorylated MBP. (B) Plot of Storm PhosphorImager units for each sample analyzed. +, present; −, absent.
Inhibition of Ad induction of IL-8 secretion by MEK inhibitor.The findings described above demonstrate that Ad7-induced activation of Erk is blocked by PD98059, an inhibitor of the kinase immediately proximal to and essential for activation of Erk. We next sought to discover whether activation of Erk was necessary for Ad7-associated induction of IL-8 secretion. A549 cells were preincubated in media with increasing doses of PD98059 (10, 25, and 50 μM) for 30 min prior to stimulation. PD98059 remained in the media for the duration of the experiments, and cells not treated with PD98059 were incubated in medium containing an equivalent amount of inhibitor solvent (DMSO). The cells were incubated for 8 h prior to harvesting of the supernatants and analysis of secreted IL-8 levels by ELISA. Ad7 increased IL-8 production ninefold over that seen in mock-infected cells (Fig.5A). Similar to the induction of Erk, Ad7-stimulated IL-8 secretion was blocked in a dose-responsive manner by PD98059. PD98059 similarly reduced IL-8 production by A549 cells responding to PMA, a potent stimulant of Erk. Analysis of IL-8 levels in 24-h supernatants (data not shown) demonstrated a comparable dose-response relationship with PD98059 treatment.
Inhibition of Ad induction of IL-8 protein by MEK inhibitor. (A) Serum-starved A549 cells preincubated with increasing doses of the MEK inhibitor PD98059 were exposed to Ad7 at an MOI of 50 or to virus-free buffer in medium plus 0.5% FBS for 8 h prior to measurement of supernatant IL-8 by ELISA in triplicate (see Materials and Methods). Cells exposed to 100 ng of PMA/ml were used as a positive control. The data are expressed as the mean + standard error of the mean of four separate experiments. (B) A549 cells were incubated as described above with the indicated amounts of U0126 and stimulated with Ad7 or PMA (100 ng/ml) for 4 h. IL-8 was measured in triplicate in supernatants by ELISA and is given as nanograms per milliliter of culture supernatant.
To confirm whether the reduction in Ad7-stimulated IL-8 production caused by PD98059 was due to inhibition of Erk, we repeated this experiment with an alternative MEK inhibitor, U0126 (14, 17). As with PD98059, U0126 reduced but did not completely abrogate IL-8 induction by Ad7 (Fig. 5B). Taken together, the results with the two MEK inhibitors establish a causal relationship between Ad7-induced Erk activation and subsequent IL-8 production.
PD98059 inhibition is reversible.To determine if inhibition of Erk activation and IL-8 induction by MEK inhibitor was due to nonspecific toxicity of PD98059, we asked whether the effects of the drug were reversible. A549 cells were incubated with 50 μM PD98059 or DMSO for 30 min. At the end of the incubation time, the cells were washed three times with PBS and resuspended in fresh medium lacking drug or solvent. The cells were then incubated for 20 min in the presence of either buffer, Ad7 at an MOI of 50, or PMA and processed for assessment of Erk activity by immunoblot analysis. As shown in Fig.6A and B, cells that were never exposed to PD98059 demonstrated fourfold and eightfold induction of Erk activity with Ad7 and PMA stimulation, respectively. Cells that had been incubated with PD98059 and washed to remove the drug regained Ad7- and PMA-inducible Erk activity, showing a five- and sevenfold activation by Ad7 and PMA, respectively. Cells that were given PD98059 after being washed with PBS had background levels of Erk induction.
PD98059 inhibition is reversible. Serum-starved A549 cells were preincubated with 50 μM PD98059 or diluent for 30 min and then washed with PBS. The cells were exposed to Ad7 at an MOI of 50 or to virus-free buffer for 20 min prior to the measurement of Erk activation or for 24 h prior to the measurement of IL-8 release (see Materials and Methods). Cells exposed to 100 ng of PMA/ml and cells exposed to PMA immunoprecipitated with NRIg were used as positive and negative controls, respectively. (A) Immunoblot of control, Ad7-infected, or PMA-stimulated and washed cell extracts stained with anti-phospho-Erk1/2 antibody (top) and same membrane stripped and reprobed with anti-Erk1/2 antibody (bottom). The Erk experiment included cells stimulated in the presence of PD98059 after being washed. (B) Activation of Erk as determined by ratio of phospho-Erk to total Erk for each immunoprecipitate. (C) IL-8 levels in supernatants of cells treated as above. The data are the mean + standard error of the mean of four separate experiments. +, present; −, absent.
In a separate set of experiments, we determined the reversibility of PD98059 inhibition of IL-8 secretion. Figure 6C shows that A549 cells never exposed to PD98059 demonstrated 5- and 11-fold increases in IL-8 with Ad7 and PMA stimulation, respectively. The fold increases in response to these stimuli were similar in cells exposed to PD98059 and washed prior to Ad7 and PMA treatment (5- and 12-fold, respectively [Fig. 6C]). These data support the notion that the effects of PD98059 on Erk and IL-8 induction are unlikely to be due to cytotoxicity but rather to a specific effect on MEK inhibition.
Erk induction is necessary for Ad7 induction of IL-8 promoter activity.Although Erk is important for the production of IL-8 in response to Ad7 infection, it is unclear which stage of IL-8 production requires Erk. Activated Erk is known to translocate into the nucleus and cause the transactivation of genes through phosphorylation of transcription factors. Since earlier studies indicated that Ad7 activated IL-8 mRNA production (7), we asked whether Erk activation was necessary for transcriptional upregulation of IL-8 mRNA. To assess IL-8 promoter activity, we utilized the −162 to +44 IL-8Luc reporter plasmid in transient-transfection assays (19). A549 cells were transfected with reporter plasmid DNA and cotransfected or not with cDNA encoding an inactive Ras mutant, N17Ras, to block stimulation of Erk (26). The transfectants were either mock infected or exposed to Ad7 for 6 h prior to lysis and measurement of reporter gene expression. As shown in Fig.7, mock-infected cells exhibited a basal level of lucerifase activity, while Ad7 infection elicited a fivefold increase in the amount of luciferase activity. In contrast, cells cotransfected with N17Ras and the IL-8 reporter plasmid displayed greatly reduced induction of luciferase by Ad7. Likewise, cells transfected with reporter plasmid and preincubated with the MEK1 inhibitor U0126 prior to infection failed to induce luciferase activity in response to Ad7 stimulation. These results confirm the necessary role of Erk in IL-8 induction by Ad7 and further indicate that the level of regulation of IL-8 production by Erk is, at least in part, at the level of transcription.
Activation of IL-8 promoter activity by Ad7 is dependent upon activation of Ras pathway. A549 cells were transfected with the Gene Porter 2 system with either IL-8Luc alone or in combination with N17Ras in fourfold excess. Cells were mock or Ad7 infected for 6 h, and luciferase activity was measured. One set of cells transfected with −162 to +44 IL-8Luc alone was preincubated with the MEK inhibitor U0126 (25 μM) for 30 min prior to mock or Ad7 infection.
Erk induction is necessary for Ad7-associated induction of endogenous IL-8 mRNA levels.We have also performed experiments to confirm that stimulation of the IL-8 gene occurs at the level of transcription and that inhibition of the Erk pathway acts at the transcriptional level. In these studies, cells were exposed to Ad7 at an MOI of 50 for 2 h in the presence or absence of 25 μM PD98059, mRNA was harvested, and IL-8 mRNA levels were determined by Northern analysis. The results (Fig. 8) demonstrate an approximately eightfold induction of endogenous IL-8 mRNA levels by Ad7. This is similar to the level of IL-8 promoter activation seen, and coupled with our previous data showing that IL-8 message stability is enhanced less than twofold by Ad7, suggests that activation of IL-8 by Ad7 occurs mostly at the level of transcription.
Induction of endogenous IL-8 mRNA levels by Ad7 is inhibited by PD98059. A549 cells were exposed to Ad7 at an MOI of 50 for 2 h in the presence or absence of 25 μM PD98059. PMA (100 ng/ml) was used as a positive control. RNA was extracted and prepared for Northern analysis as described in Materials and Methods. (A) RNA blot probed for IL-8 mRNA. (B) RNA gel of the blot in panel A stained with ethidium bromide. (C) IL-8 mRNA levels expressed as a ratio of IL-8 mRNA to rRNA visualized for each condition.
Preincubation of Ad7-stimulated cells with PD98059 inhibited this induction by 90%. Thus, inhibition of Erk pathway activation by PD98059 likely acts at the transcriptional level by inhibiting Erk-modulated IL-8 promoter activation. This prevents the enhancement of endogenous IL-8 mRNA levels by Ad7.
DISCUSSION
Host cells respond to acute infectious agents by activation of cellular signaling pathways and by the production of inflammatory mediators necessary to eliminate the pathogen. We sought to determine whether induction of the inflammatory mediator IL-8 by Ad infection of human epithelial cells involves the Erk1/2 pathway. Our findings presented here demonstrate that IL-8 induction requires activation of Erk. First, Ad7 infection of A549 human lung epithelial cells provoked a rapid stimulation of Erk enzymatic activity, assessed by two independent methods. Second, increasing inhibition of Ad7-induced Erk activation by the MEK inhibitor PD98059 or U0126 correlated with increased inhibition of Ad7-induced IL-8. Additionally, Ad7 infection of A549 cells stimulated the IL-8 promoter, indicating that Ad7 infection elicits nascent mRNA encoding IL-8. Most importantly, inhibition of the Erk signaling pathway in A549 cells using a dominant-negative Ras in transient-transfection assays resulted in abrogation of IL-8 promoter induction by Ad7.
These findings are consistent with a model in which Ad7 stimulates the Ras pathway in host cells, resulting in the activation of Erk. Activated Erk stimulates the IL-8 promoter to elicit nascent IL-8 mRNA and protein. The IL-8 protein is secreted by virus-infected cells to produce an inflammatory response. The data provide insight into the cellular mechanisms of host response to wild-type Ad infection known to result in robust inflammation. Furthermore, the results suggest possible mechanisms for inhibiting the inflammatory response to the virus during gene therapy protocols involving Ad vectors.
Although it is unclear how the virus initiates this host response, we hypothesize that activation of Erk, with subsequent induction of IL-8 and inflammation, is an event driven by Ad7 interaction with host receptors. This notion is supported by the rapid kinetics of Erk phosphorylation, which peaks within 10 min of Ad7 contact with cells. Furthermore, the hypothesis is consistent with the findings of other groups studying the induction of cytokines by recombinant Ad vectors. In these examples, induction of the cytokine IP-10 could be measured in cells infected with Ad in which the viral genome had been inactivated by radiation (8, 34). The initial steps of Ad infection involve attachment of the globular head of the Ad fiber to a cellular receptor, followed by interaction of the penton base of the virus with a host cell integrin receptor through an arginine-glycine-aspartic acid (RGD) motif (40). Subgroup B viruses, such as Ad7, have been reported to interact with the same αvβ3 and αvβ5integrins utilized by the other Ad serotypes (32). However, Ad7 and other subgroup B viruses do not utilize the coxsackievirus and adenovirus receptor utilized by most other Ad serotypes as the fiber receptor (37). The use of alternate fiber receptors may influence intracellular trafficking of Ad serotypes. Thus, Ad5f7 chimeras, containing the Ad5 capsid with Ad7 fibers, progress through the cytoplasm in a manner similar to that of the Ad7 virus and distinct from that of the Ad5 virus (33). The induction of signal transduction pathways and inflammatory mediators by Ad7 may be the result of stimulation of receptors distinct from those used by other Ad subgroups.
Although our data indicate that induction of the Erk pathway is essential for Ad7-mediated induction of IL-8 mRNA, it is possible that other signal transduction pathways are also involved. This is suggested by the fact that levels of the MEK inhibitors that completely eliminate Erk activation were not entirely sufficient to eliminate IL-8 secretion. Because Ad internalizes through an integrin ligation-mediated process, it is likely that signal transduction events associated with integrin function are activated during Ad internalization. For example, Ad2 induces the activation of phosphatidylinositol 3-kinase, which accompanies viral endocytosis by human epithelial cells (28). Ad infection likewise activates an Src kinase, p125 focal adhesion kinase, and p130Cas (29). These proteins form a signaling complex that presumably leads to activation of additional signal transduction pathways besides one involving Ras/Raf/MEK and Erk (29). Additional signal transduction pathways induced by Ad7 might similarly contribute to the production of inflammatory mediators by host cells. IL-8 induction by other pathogens has been shown to require activation of the p38 MAPK (2, 18, 31). We are presently investigating the role of other pathways, including p38 MAPK, in the induction of the IL-8 promoter by Ad7.
The results reported here reveal that infection of lung epithelial cells with Ad7 results in the production of IL-8, and the attendant lung inflammation, through a signal transduction process that involves the Ras/Raf/MEK/Erk pathway. Erk appears to regulate IL-8 production at the level of transcription. It seems unlikely that Ad7 stimulates secretion of preformed IL-8 in host cells, since IL-8 mRNA levels in uninfected cells are much lower than those seen after Ad7 infection (7). Furthermore, our finding that Ad7 stimulates the IL-8 promoter in A549 cells is not consistent with a model in which host cells secrete preformed IL-8. However, none of our experiments specifically address secretion of preformed IL-8. An additional possibility is that virus internalization by host cells and not receptor triggering per se is the essential event leading to IL-8 production. The findings reported here cannot eliminate this possibility, and we are investigating the issue. Differences in fiber receptors on host cells for different Ad serotypes may explain why wild-type Ad5 does not induce clinically significant inflammation in the lower respiratory tract (7). However, Erk activation and IL-8 production were reported when epithelial cells were infected with E1-, E3-, and E4-deficient Ad5 (10), but IL-8 is not induced upon infection with wild-type Ad5 (7). Hence, the use of different receptors alone does not account for differences in host responses upon infection of host cells by the various Ad serotypes. Deciphering the role of Ad receptor interactions, signal transduction pathways, and induction of the inflammatory response will be especially important for the future design of gene therapy vectors and treatment protocols for Ad infections.
ACKNOWLEDGMENTS
J. P. Metcalf is supported by NIH K08-HL03106 from the NHLBI, an American Heart Association Heartland Affiliate Grant, and a Presbyterian Health Foundation Grant. K. M. Coggeshall is a Scholar of the Leukemia & Lymphoma Society (formerly Leukemia Society of America).
We acknowledge the assistance of Gary Kinasewitz and Linda Thompson for review of the manuscript.
FOOTNOTES
- Received 16 January 2001.
- Accepted 13 April 2001.
- Copyright © 2001 American Society for Microbiology
