Abstract
Lung tissue from patients with emphysema and airway obstruction carries excess adenoviral E1A DNA that is expressed as protein in airway surface epithelium and is associated with an increased inflammatory response. To examine mechanisms by which latent adenoviral infection might amplify the inflammatory process, we transfected primary human bronchial epithelial (HBE) cells from three separate patients undergoing lung resection so that they stably expressed adenovirus E1A. Lipopolysaccharide stimulation of the E1A-transfected HBE cells increased intercellular adhesion molecule-1 and interleukin-8 mRNA and protein expression compared with control cells from the same patient. It also induced greater intercellular adhesion molecule-1 promoter activity and greater nuclear factor-κB binding activity of nuclear extracts in E1A transfectants than controls. E1A-positive transfectants constitutively expressed transforming growth factor-β1 mRNA and protein, whereas this expression was either very low or not detected in control cells. We conclude that adenoviral E1A tranfection transforms primary HBE cells and upregulates their production of mediators that are clinically relevant to the pathogenesis of chronic obstructive pulmonary disease.
Everyone who smokes develops lung inflammation (1), but only a susceptible minority develops chronic airway obstruction (2). A recent report by Retamales and colleagues suggests that the cigarette smoke-induced inflammatory process that underlies emphysematous destruction of the lung in chronic obstructive pulmonary disease (COPD) is amplified in smokers with advanced disease compared with those with similar smoking histories and preserved lung structure and function (3). This enhanced response to a similar degree of stimulation was associated with evidence of latent adenoviral infection of the alveolar surface epithelium (3).
Studies of an animal model of human adenovirus 5 infection in the guinea pig have shown that deposition of the virus in the nasal cavity is followed by viral replication in the lung and persistence of the viral E1A gene with expression of its protein in lung epithelial cells long after viral replication stops (4). Animals with this form of latent adenoviral infection develop an excessive inflammatory reaction following a single exposure to cigarette smoke (5) and excess lung inflammation and emphysematous lung destruction following chronic cigarette smoke exposure (6). Our working hypothesis is that an acute adenoviral infection is followed by the persistence of viral genes in lung epithelial cells where expression of the transactivating adenoviral E1A gene results in transformation of the infected cell with overexpression of host inflammatory genes. Work reported from several laboratories has established that the E1A viral protein can cause cells to enter their replicative cycle (reviewed in 7) and function as an ubiquitous enhancer of a large number of host genes by binding to the DNA binding sites of several transcription factors (reviewed in 8). These studies were undertaken to determine whether transfection of the E1A gene into primary human bronchial epithelial (HBE) cells results in an increased expression of inflammatory mediators that have been implicated in the pathogenesis of COPD.
Studies have shown that COPD is associated with an elevation of induced sputum concentrations of interleukin (IL)-8 (9) and that intercellular adhesion molecule (ICAM)-1 is abundantly present in the lung epithelium of resected lung tissue from smokers (10, 11). ICAM-1 has also been reported to be upregulated in the epithelium of patients with obstructive bronchitis (11) and is overexpressed in association with E1A protein as emphysematous lung destruction progresses (3). This study tests the hypothesis that the adenoviral E1A gene upregulates ICAM-1 and IL-8 mRNA expression in HBE cells by increasing the promoter activity of these genes. We studied the transcription factors nuclear factor (NF)-κB and activator protein (AP)-1 because they have binding sites in the regulatory regions of both genes (12, 13). The effect of E1A transfection on the expression of transforming growth factor (TGF)-β was also examined because its mRNA expression is increased in airway epithelium from patients with COPD (14), and it has been implicated in the tissue remodeling process that results in airway obstruction (14). Furthermore, its ability to induce nonneoplastic cells to express a transformed phenotype (15) could explain how this viral gene is able to persist in airway epithelium following an acute infection.
Lung tissue was obtained from three patients who required resection for a bronchogenic carcinoma (Table 1)
Case No. | Age (yr) | Sex | Resected Lung | Smoking History (pack-yr) | Lung Function | HBE Cells | ||
|---|---|---|---|---|---|---|---|---|
| FEV1 (%) | FVC (%) | |||||||
| 3,491 | 76 | Male | Left upper lobe | 40 | 90 | 92 | HC34 and HA34 | |
| 3,501 | 64 | Male | Left upper lobe | 35 | 115 | 113 | HC35, HA35-1, and HA35-2 | |
| 5,726 | 57 | Male | Left upper lobe | 44 | 91 | 102 | HC57 and HA57 | |
HBE cells were grown in bronchial epithelial cell growth medium (Clonetics, San Diego, CA). A549 cells, E4, and C5, stably transfected with adenovirus E1A or control plasmid, respectively (8), were grown in Eagle's minimum essential medium (GIBCO BRL, Gaithersburg, MD) supplemented with 10% fetal bovine serum and maintained under constant selection with 280 μg/ml G418 (GIBCO BRL). Peripheral blood monocytes were cultured in RPMI 1640 medium (GIBCO BRL) with 10% fetal bovine serum.
Plasmid pE1Aneo containing the adenovirus 5 E1A gene alone has been described (8). The pXC-15 plasmid carrying the entire coding region of the E1A plus E1B genes of adenovirus 5 was digested with EcoRI/ApaI, and this fragment from nucleotides 1 to 4,619 was ligated with pSV2neo (Clontech Laboratories, Palo Alto, CA) and called pE1A/E1Bneo.
When 60% to 90% confluent, primary HBE cells from passages 1 to 3 plated on six-well plates were transfected using the Lipofectamine Reagent (GIBCO BRL). G418 (GIBCO BRL)-resistant colonies were isolated by limiting dilution in 96-well plates. All resistant clones were maintained under constant selection with 100 μg/ml of G418 except for one, which was further cultured without G418.
Polymerase chain reaction (PCR) (5), immunocytochemistry (8), Northern analysis (17) of total RNA extracted with TRIzol (GIBCO BRL), and enzyme-linked immunosorbent assay (8, 17) were performed as described.
HBE cells were transfected with plasmid DNA using diethylaminoethyl-dextran. Plasmid DNA was either pBS-CAT-P, which carries the ICAM-1 5′-flanking region (12) linked to the chloramphenicol acetyltransferase (CAT) coding region (18) or pCAT-control (Promega, Madison, WI). One day later, cells remained unstimulated or were stimulated with lipopolysaccharide (LPS) (10 μg/ml) (Sigma Chemical Co., St. Louis, MO) or interferon (IFN)-γ (100 U/ml) (GIBCO BRL) for 24 hours before analysis by the CAT assay (19). To correct for differences in transfection efficiency between untransfected and pE1A/E1Bneo-transfected cells, the ICAM-1 promoter activity was expressed as a ratio of CAT activity in pBS-CAT-P–transfected cells to that found in the same cells transfected with pCAT-control.
Nuclear protein extraction and electrophoretic mobility shift assay were described previously (13).
An analysis of variance compared the ratio of the intensity of toll-like receptor-2 (TLR2) mRNA band corrected by that of the corresponding glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA from E1A-transfected A549 cells to the ratio from A549 cells transfected with control plasmid. Data from the ICAM-1 promoter assay were compared using the Student's t test; p values were corrected for multiple comparisons by the ranked Bonferroni method as described (8), and p < 0.05 was considered significant.
Transfections of HBE cells with pE1Aneo were repeated eight times with lipofection, but no stable E1A-expressing transfectants were obtained. Transfections with pE1A/E1Bneo produced G418-resistant colonies within 3 to 4 weeks, and the four single colonies of resistant cells isolated from HBE cells from the three donors are referred to as HA34, HA35-1, HA35-2, and HA57 (Table 1). Unlike the primary HBE cells shown in Figure 1A
Figure 1. Morphology of primary HBE cells and pE1A/E1Bneo transfectants by light microscopy (left panels) and electron microscopy (right panels). (A) Untransfected primary HBE cells, HC57, showing normal morphology. (B) HA35-2 cells 10 months after transfection with pE1A/E1Bneo by lipofection are tightly clustered. (C) HA57 cells 4 months after transfection with pE1A/E1Bneo show morphologic characteristics similar to the primary cells in (A). (D) By electron microscopy an untransfected HC57 cell grown on a polycarbonate filter was unflattened but was not columnar, possibly because of the lack of an extracellular matrix substrate. It had few apical microvilli and filament bundles (arrowheads) around the nucleus. (E) Electron microscopy of E1A-expressing HA57 grown on a polycarbonate filter shows a columnar cell with apical microvilli and junctional complexes (arrows). (F) Higher magnification of an apical junctional complex (arrow) between two HA57 cells. In A to C, bar = 50 μm; in D to F, bar = 1 μm.
[More] [Minimize]Electron microscopy, used to investigate the morphology of the epithelial cells, showed that untransfected HBE cells grown on polycarbonate filters had a squamoid shape compared with their E1A-expressing counterparts that had a columnar shape (Figures 1D and 1E). Both cell types had vacuolated cytoplasmic vesicles and apical microvilli. The untransfected controls contained filament bundles (Figure 1D), and E1A-expressing cells expressed junctional complexes (Figures 1E and 1F).
Verification that the G418-resistant cells carried the adenovirus E1A DNA was made by amplification of this viral DNA by PCR. E1A DNA was amplified from DNA extracted from all four pE1A/E1Bneo transfectants but not from the untransfected HBE cells (Figure 2A)
Figure 2. Detection of E1A DNA by PCR, E1A protein by immunocytochemistry, and E1A and E1B mRNA expression by Northern analysis. (A) E1A DNA of adenovirus 5 was amplified by PCR on DNA extracted from pE1A/E1Bneo transfectants, HA34, HA35-1, HA35-2, and HA57 (right lanes), but not from the corresponding untransfected HBE cells from the same patients, HC34, HC35, and HC57 (left lanes). The 486-bp E1A PCR product was detected by ethidium bromide staining after electrophoresis in a 1% agarose gel. (B) HA57 cells fixed with formalin and acetone were stained with monoclonal antibodies to E1A (left) or mouse IgG (control antibody) (right) before detection of antibody binding by the alkaline phosphatase antialkaline phosphatase (APAAP) method. Bar = 100 μm. (C) Autoradiograms of a Northern blot of RNA from A549 cells infected with adenovirus 5 (Ad5-infected A549); untransfected HBE cells, HC34, HC35, and HC57; and pE1A/E1Bneo transfected HBE cells, HA34, HA35-1, HA35-2, and HA57, that was probed with either E1A DNA (upper panel), E1B DNA with arrowheads indicating the 13S and 22S transcripts (middle panel), or GAPDH cDNA as an internal control (lower panel) are shown. The same filter was hybridized with each probe after removal of the previous one in boiling water. For the autoradiogram of the E1B Northern, film exposure for the Ad5-infected A549 lane was 6 hours and for the remaining lanes was 24 hours. Arrows indicate positions of the 18S and 28S ribosomal RNA bands in the corresponding ethidium bromide stained gel.
[More] [Minimize]To investigate whether the transfected HBE cells express the E1A and E1B proteins, cells were stained with the respective antibodies. Three of the four cell lines established (HA34, HA35-1, and HA57) showed nuclear staining for E1A (Figure 2B) similar to adenovirus 5–infected A549 cells (referenced in 8), and one (HA35-2) did not. E1B 19K protein was not detected in the transfectants, although strong cytoplasmic staining for E1B 19K was observed in adenovirus 5–infected A549 cells (data not shown). Cytoplasmic staining for E1B 55K protein was seen in only 0.1 to 2.3% of E1A-positive transfected cells (HA34, HA35-1, and HA57), whereas none was seen in adenovirus 5–infected A549 cells or in HA35-2 (data not shown). All three untransfected HBE cells were negative for E1A and E1B proteins, whereas they and their pE1A/E1Bneo-transfected counterparts were positive for cytokeratin but were negative for CD14 (data not shown). Peripheral blood monocytes used as a control for CD14 expression were strongly positive for this LPS receptor (data not shown).
Expression of these viral genes was further investigated at the mRNA level. By Northern analysis, E1A mRNA was detected in pE1A/E1Bneo-transfected HBE cells (HA34, HA35-1, and HA57) as well as in adenovirus 5–infected A549 cells (Figure 2C). Untransfected HBE cells and HA35-2, the pE1A/E1Bneo-transfected cell line not staining for E1A protein, did not express E1A mRNA. Compared with A549 cells infected with adenovirus 5 that produce two E1A mRNA transcripts, most likely the 12S and 13S, only the larger was found in E1A-positive HBE cells.
E1A-positive pE1A/E1Bneo-transfected HBE cells expressed much lower levels of E1B 13S and 22S mRNAs compared with adenovirus 5–infected A549 cells (Figure 2C). Also, the 22S transcript predominated in the three E1A-positive HBE cells, whereas the 13S mRNA was more prominent in the adenovirus 5–infected A549 cells. Untransfected controls and HA35-2 did not express E1B mRNAs. GAPDH mRNA levels were similar in all cells tested. Most important, our results demonstrate that the E1A gene transfected into HBE cells is expressed as mRNA and protein.
Northern analysis showed that all E1A-expressing transfectants constitutively expressed TGF-β mRNA with little or no change after stimulation (Figure 3)
Figure 3. TGF-β1 mRNA expression. Control HBE cells (E1A−) HC34, HC35 and HC57, and pE1A/E1Bneo transfectants expressing E1A (E1A+), HA34, HA35-1, and HA57 and not expressing E1A (E1A−), HA35-2 were incubated in medium alone (−) or in medium with LPS (10 μg/ml), TNF-α (100 U/ml), or IFN-γ (100 U/ml) for 6 hours. RNA was extracted from the cells and autoradiograms of the resulting Northern blots probed with cDNAs of TGF-β1 or GAPDH as an internal control are shown.
[More] [Minimize]Without stimulation, ICAM-1 mRNA levels were very low or not detected in all four pE1A/E1Bneo transfectants and the corresponding control cells (Figure 4A)
Figure 4. ICAM-1 and IL-8 mRNA expression. RNA for Northern blotting was extracted from the same cells with the same stimulation conditions as in Figure 3. Autoradiograms of the resulting Northern blots probed with either (A) ICAM-1 or (B) IL-8 cDNA are shown.
[More] [Minimize]Basal levels of IL-8 mRNA expression in untransfected control cells varied from minimal expression in HC57 to moderate expression in HC35, whereas in all pE1A/E1Bneo transfectants, this mRNA was not detected (Figure 4B). LPS stimulation dramatically enhanced expression in E1A-expressing cells, whereas its effect on untransfected HBE cells and on HA35-2, the transfectant not expressing E1A, was moderate. TNF-α consistently increased IL-8 mRNA expression above basal levels in all cells, but TNF-α stimulation of pE1A/E1Bneo-transfected cells was not as effective as LPS. IFN-γ affected IL-8 expression minimally in both controls and transfectants.
Analysis of inflammatory mediator protein expression by ELISA showed that when IL-8 expression in response to LPS stimulation was compared with the unstimulated state, the ratio increased from 1.3 ± 0.8 in E1A-negative HC57 cells to 37 ± 11 in the E1A HA57-positive cells (average ± SD, n = 6, p < 0.00004). Similarly, for ICAM-1 protein expression, the ratio increased from 0.3 ± 0.5 in the E1A-negative cells to 12 ± 4 in the positive cells (average ± SD, n = 6, p < 0.0002). TGF-β protein was also increased in E1A-positive HBE cells compared with controls (20; data not shown).
Because HBE cells did not express CD14, the expression of a second LPS receptor, TLR2, was investigated. The TLR2 cDNA probe (21) prepared from a plasmid from Genentech Inc. (San Francisco, CA) showed that untransfected (HC57) and E1A-positive (HA57) HBE cells expressed TLR2 mRNA at low levels compared with peripheral blood monocytes (Figure 5A)
Figure 5. TLR2 mRNA expression and ICAM-1 promoter activity. (A) RNA was extracted from untreated peripheral blood monocytes (PBM) and, after incubation without (−) or with LPS (10 μg/ml) for 6 hours, from untransfected HC57 cells (E1A−) and the corresponding E1A-expressing HA57 cells (E1A+) and from A549 cells C5 (E1A−) and E4 (E1A+). Autoradiograms of the resulting Northern blots probed with cDNAs of TLR2 or internal control GAPDH are shown. (B) E1A-expressing transfectant HA57 (black bar) and the corresponding control HC57 (white bar) were transiently transfected with 1 μg/ml pBS-CAT-P or pCAT-control. Cells were left unstimulated (NS) or stimulated with LPS (10 μg/ml) or IFN-γ (100 U/ml) for 24 hours and then analyzed by the CAT assay. CAT activity of the pBS-CAT-P–transfected cells was normalized to that found in the same cells transfected with pCAT-control. Data are expressed as means ± SEM of three independent experiments. *p < 0.05.
[More] [Minimize]In contrast, A549 cells transfected with E1A expressed TLR2 mRNA, whereas those transfected with control plasmid did not (Figure 5A). Densitometric analysis showed that although LPS stimulation did not affect this expression, the ratios of the density of the TLR2 mRNA to that of the corresponding GAPDH mRNA was significantly higher in E1A-positive A549 cells (0.033 ± 0.007) than in controls (0.009 ± 0.008) (average ± SD, n = 6, p < 0.002).
To determine whether adenovirus E1A modulates the expression of specific inflammatory mediators by affecting the promoters of these genes, the ICAM-1 promoter linked to a CAT reporter was transfected into the HBE cells. ICAM-1 promoter activity was increased over basal levels by LPS in both untransfected (HC57) and E1A-expressing (HA57) cells (Figure 5B). This LPS-induced activity was higher in E1A-positive HA57 cells than in HC57. IFN-γ stimulation did not affect promoter activity in these cells.
Further support for E1A's modulation of inflammatory mediator gene expression was sought by studying the binding activity of transcription factors known to regulate both ICAM-1 and IL-8 genes. When the electrophoretic mobility shift assay using the oligonucleotide specifying the NF-κB binding sequence was applied to nuclear extracts from cells grown in medium alone, specific DNA–protein complexes were detected at low levels in E1A-positive transfectant (HA57) and the untransfected cells (HC57) (Figure 6A)
Figure 6. Detection of NF-κB binding activity in untransfected HC57 cells (E1A−) and E1A-positive transfectant HA57 (E1A+) by electrophoretic mobility shift assay using the NF-κB oligonucleotide. (A) After HC57 and HA57 were incubated for 2 hours in medium alone (−) or medium with LPS (10 μg/ml), TNF-α (100 U/ml), or IFN-γ (100 U/ml), nuclear extracts from these cells were mixed with radiolabeled NF-κB oligonucleotide. An autoradiogram detecting NF-κB binding activity is shown. The positions of specific protein–DNA complexes (I, II, and III) and nonspecific complexes (NS) are indicated by arrows. (B) After HC57 and HA57 were incubated with TNF-α and HA57 with LPS as in (A), their nuclear extracts were pretreated with a 50-fold excess competitor unlabeled NF-κB oligonucleotide or unrelated AP-1 oligonucleotide before radiolabeled NF-κB oligonucleotide was added. (C) After HA57 was incubated with LPS as in (A), nuclear extracts were either left untreated (−) or were treated with antiserum against p50, RelA, or unrelated proteins (anti-Un) before the radiolabeled oligonucleotide was added. Positions of specific complexes (I, II, and III) and NS indicated by arrows correspond to those shown in (A) and (B), whereas those of the supershifted bands are indicated by arrowheads.
[More] [Minimize]When basal levels of AP-1 binding activity were tested both transfectant HA57 and control HC57 cells demonstrated binding (Figure 7)
Figure 7. Detection of AP-1 binding activity in the nuclei of untransfected HC57 (E1A−) and corresponding E1A-positive HA57 transfectant (E1A+) by electrophoretic mobility shift assay with the AP-1 consensus oligonucleotide. An autoradiogram of the electrophoretic mobility shift assay on nuclear extracts from these cells after incubation in medium alone (−), LPS, TNF-α, or IFN-γ as described in Figure 6, together with a competition assay to determine specific AP-1 binding activity in the right two lanes, is shown. For the latter, nuclear extracts from HA57 cells cultured for 2 hours with medium alone were pretreated with a 50-fold excess competitor unlabeled AP-1–specific oligonucleotide or unrelated NF-κB oligonucleotide before incubation with the radiolabeled AP-1 oligonucleotide. The position of the specific AP-1-DNA complex is indicated by an arrow.
[More] [Minimize]Results from both the ICAM-1 promoter assay and electrophoretic mobility shift assay demonstrate that adenovirus E1A alters the response of the ICAM-1 and IL-8 genes to LPS stimulation through mechanisms involving the 5′ regulatory sequences of these genes.
These studies provide new information concerning the excess production of ICAM-1, IL-8, and TGF-β by primary HBE cells transfected with the adenoviral E1A gene. These observations are relevant to the hypothesis that lung epithelial cells that express the viral E1A gene are capable of persisting in the airway epithelium and driving an overproduction of mediators that have been implicated in the pathogenesis of COPD. Transformation of the primary HBE cells reported here and of human embryonic kidney cells used to develop the 293 cells (22) required adenovirus E1B in addition to E1A. This requirement may be related to the capacity of E1B to neutralize apoptosis by several mechanisms (23–25). This suggests that E1B is necessary for transformation of primary human cells by adenovirus E1A but either not necessary to maintain the transformed state or required at levels too low for immunocytochemical detection.
Results from electron microscopic and immunocytochemical analysis of the transfected HBE cells showed that many airway epithelial characteristics were retained. They have microvilli on their apical surface, vacuolated vesicles in the cytoplasm, junctional complexes between adjacent cells, and a columnar shape when grown on a polycarbonate surface. The transfected cells were also positive for cytokeratin. The maintenance of these epithelial characteristics suggests that a population of these transformed cells could remain in the airway epithelium following infection.
The enhanced expression of ICAM-1 and IL-8 mRNAs in E1A-expressing HBE cells indicates that adenovirus E1A can modulate the transcription of inflammatory mediator genes in response to LPS stimulation. In contrast, untransfected HBE cells from the same patient source did not respond to LPS with respect to ICAM-1 expression and showed only a moderate response with respect to IL-8. Additional support that E1A modulates ICAM-1 gene transcription comes from results of the CAT reporter assay where the presence of E1A resulted in an increase in the induction of the ICAM-1 promoter by LPS. These differences between the E1A-positive HBE cells and their controls were most likely not a consequence of transformation or immortalization because HA35-2 that was transfected but did not express E1A responded to LPS in the same manner as the untransfected HBE cells.
Activated NF-κB is necessary for the transcription of both ICAM-1 (12) and IL-8 (13) genes, and our data showed that increases in NF-κB binding in response to LPS stimulation of HBE cells required E1A expression. The activation of NF-κB and the enhanced response of these genes are not explained by an altered expression of the LPS receptors, CD14 or TLR2, because CD14 was not expressed by either untransfected or E1A-positive HBE cells, and TLR2 mRNA was increased only marginally in the presence of E1A. The E1A-induced changes in gene regulation in HBE cells are similar to those found in A549 cells expressing E1A (8, 13, 17), but the mechanism of induction may be different because only E1A-expressing A549 cells express TLR2, whereas both E1A-positive and -negative HBE cells express this receptor. An alternative target of E1A modulation is TLR4, another member of this family of genes that is expressed by HBE cells (26).
TNF-α signals through a pathway different from LPS (27), but the electrophoretic mobility shift assays confirmed that both pathways involve the activation of NF-κB in HBE cells. ICAM-1 expression was strongly stimulated by both TNF-α and LPS in E1A-expressing HBE cells but not when E1A was absent. In contrast, IL-8 expression was more strongly stimulated by TNF-α than by LPS in the absence of E1A, whereas LPS was more effective when E1A was present. Whether the number of NF-κB binding sites in the 5′ regulatory region of these genes, ICAM- with two (12) and IL-8 with one (references in 13), accounts for the differential effect of E1A on the response to the two inflammatory stimuli remains to be determined.
The strong induction of ICAM-1 mRNA by IFN-γ in untransfected HBE cells and a general dampening of this response by E1A are compatible with reports in other lung epithelial cells (8, 28–30). The inhibition by E1A can be attributed to a block in the phosphorylation and activation of the STAT transcription factor that binds to the IFN-γ response element of the ICAM-1 gene in airway epithelial cells (30). This repression of the IFN signal transduction pathway is proposed as a strategy by which adenovirus escapes immune surveillance (31). The absence of an IFN-γ response in HA35-2, the E1A/E1B transfectant not expressing E1A, remains unexplained. The weak induction of IL-8 mRNA by IFN-γ in untransfected HBE cells, on the other hand, can be attributed to the lack of an IFN response element in the regulatory region of the IL-8 gene (see references in 13).
Constitutive expression of TGF-β1 mRNA in our transfected HBE cells, where the larger 13S E1A mRNA is expressed but not the 12S transcript, is consistent with the finding that the TGF-β1 promoter activity is repressed by the 12S but not the 13S product of the E1A gene (32). Although TGF-β may be necessary for the maintenance of these cells in the transformed state, this growth factor is also known to be a potent growth inhibitor of normal HBE cells (33). Adenovirus E1A could maintain HBE cells in a transformed state while allowing continued growth of these TGF-β1–expressing cells by its binding to the retinoblastoma gene product and reversing growth suppression (reviewed in 7).
The increased expression of inflammatory mediators, TGF-β, ICAM-1, and IL-8, by E1A-transformed HBE cells in culture is relevant to the pathogenesis of COPD. For TGF-β, the degree of mRNA overexpression reported in the airway epithelium of smokers with COPD correlated with the severity of the airway obstruction (14). TGF-β1 is a fibrogenic growth factor thought to be involved in the airway remodeling found in chronic inflammatory disorders such as COPD and asthma (34, 35). ICAM-1 and IL-8 are mediators of neutrophil chemotaxis and adhesion, respectively, that are increased in COPD patients (3, 36), and their increase accounts for the higher neutrophil counts in the lavage fluid from these patients (37). Therefore, the finding that adenovirus E1A markedly enhances the expression of these mediators in HBE cells suggests mechanisms for the persistence of a cell line that could promote both airways remodeling and neutrophilic airways disease in COPD.
The authors thank Dr. Frank Graham, McMaster University, for the pE1Aneo plasmid; Dr. Tom Shenk, Princeton University, for pXC-15; Dr. S. W. Caughman, Emory University, for pBS-CAT-P; Genentech for TLR2 cDNA; and Drs. Peter Pare and Naoto Keicho for critical reading of the manuscript.
Supported by MRC 7246, the National Centres of Excellence for Respiratory Health, and the British Columbia Lung Association. Y. H. is a recipient of an MRC/PMAC-Astra Pharma Respiratory Medicine Fellowship.
| 1. | Niewoehner DE, Kleinerman J, Rice DB. Pathologic changes in the peripheral airways of young cigarette smokers. N Engl J Med 1974;291: 755–758. |
| 2. | Fletcher C, Peto R. A natural history of chronic airflow obstruction. BMJ 1977;1:1645–1648. |
| 3. | Retamales I, Elliott WM, Meshi B, Coxson HO, Pare PD, Sciurba FC, Rogers RM, Hayashi S, Hogg JC. The amplification of inflammation in emphysema and its association with latent adenoviral infection. Am J Respir Crit Care Med 2001;164:469–473. |
| 4. | Vitalis TV, Keicho N, Itabashi S, Hayashi S, Hogg JC. An animal model of adenovirus infection. Am J Respir Cell Mol Biol 1996;14:225–231. |
| 5. | Vitalis TZ, Kern I, Croome A, Hayashi S, Hogg JC. The effect of adenovirus 5 infection on cigarette smoke-induced lung inflammation. Eur Respir J 1998;1:664–669. |
| 6. | Meshi B, Vitalis T, Ionescu D, Hayashi S, Hogg JC. Emphysematous lung destruction by cigarette smoke: effect of latent adenovirus 5 infection and retinoic acid treatment. Am J Respir Cell Mol Biol 2002; 26:52–57. |
| 7. | de Groot RP, Kranenburg O, de Wit L, van den Eijnden-van Raaij J, Mummery C, van der Eb AJ, Zantema A. Adenovirus E1A antagonizes both negative and positive growth signals elicited by transforming growth factor β1. Cell Growth Differ 1995;6:531–540. |
| 8. | Keicho N, Elliott WM, Hogg JC, Hayashi S. Adenovirus E1A gene dysregulates ICAM-1 expression in transformed pulmonary epithelial cells. Am J Respir Cell Mol Biol 1997;16:23–30. |
| 9. | Yamamoto C, Yoneda T, Yoshikawa M, Fu A, Tokuyama T, Tsukaguchi K, Narita N. Airway inflammation in COPD assessed by sputum levels of interleukin-8. Chest 1997;112:505–510. |
| 10. | Di Stefano A, Maestrelli P, Roggeri A, Turato G, Calabro S, Potena A, Mapp CE, Ciaccia A, Covacev L, Fabbri LM, et al. Upregulation of adhesion molecules in the bronchial mucosa of subjects with chronic obstructive bronchitis. Am J Respir Crit Care Med 1994;149:803–810. |
| 11. | Gonzalez S, Hards J, van Eeden S, Hogg JC. The expression of adhesion molecules in cigarette smoke-induced airways obstruction. Eur Respir J 1996;9:1995–2001. |
| 12. | Stratowa C, Audette M. Transcriptional regulation of the human intercellular adhesion molecule-1 gene: a short overview. Immunobiology 1995;193:293–304. |
| 13. | Keicho N, Higashimoto Y, Bondy GP, Elliott WM, Hogg JC, Hayashi S. Endotoxin-specific NF-κB activation in pulmonary epithelial cells haboring adenovirus E1A. Am J Physiol 1999;277:L523–L532. |
| 14. | Takizawa H, Tanaka M, Takami K, Ohtoshi T, Ito K, Satoh M, Okada Y, Yamasawa F, Nakahara K, Umeda A. Increased expression of transforming growth factor-1 in small airway epithelium from tobacco smokers and patients with chronic obstructive pulmonary disease (COPD). Am J Respir Crit Care Med 2001;163:1476–1483. |
| 15. | Roberts AB, Anzano MA, Wakefield LM, Roche NS, Stern DF, Sporn MB. Type β transforming growth factor: a bifunctional regulator of cellular growth. Proc Natl Acad Sci USA 1985;82:119–123. |
| 16. | Gruenert DC, Basbaum CB, Widdicombe JH. Long-term culture of normal and cystic fibrosis epithelial cells grown under serum-free conditions. In Vitro Cell Dev Biol 1990;26:411–418. |
| 17. | Keicho N, Elliott WM, Hogg JC, Hayashi JC. Adenovirus E1A upregulates interleukin-8 expression induced by endotoxin in pulmonary epithelial cells. Am J Physiol 1997;272:L1046–L1052. |
| 18. | Degitz K, Li LJ, Caughman SW. Cloning and characterization of the 5′-transcriptional regulatory region of the human intercellular adhesion molecule 1 gene. J Biol Chem 1991;266:14024–14030. |
| 19. | Gorman CM, Moffat LF, Howard BH. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 1982;2:1044–1051. |
| 20. | Ogawa E, Higashimoto Y, Hogg JC, Hayashi S. The role of adenovirus E1A in airway remodeling [abstract]. Am J Respir Crit Care Med 2001;163:A736. |
| 21. | Yang RB, Mark MR, Gray A, Huang A, Xie MH, Zhang M, Goddard A, Wood WI, Gurney AL, Godowski PJ. Toll-like receptor-2 mediates lipopolysaccharide-induced cellular signalling. Nature 1998;395:284–288. |
| 22. | Graham FL, Smiley J, Russell WC, Nairn R. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol 1977;36:59–74. |
| 23. | Debbas M, White E. Wild-type p53 mediates apoptosis by E1A, which is inhibited by E1B. Genes Dev 1993;7:546–554. |
| 24. | Sarnow P, Ho YS, Williams J, Levine AJ. Adenovirus E1b-58kd tumor antigen and SV40 large tumor antigen are physically associated with the same 54 kd cellular protein in transformed cells. Cell 1982;28:387–394. |
| 25. | Shen Y, Shenk T. Relief of p53-mediated transcriptional repression by the adenovirus E1B 19-kDA protein or the cellular Bcl-2 protein. Proc Natl Acad Sci USA 1994;91:8940–8944. |
| 26. | Reddy SPM, Kleeberger SR, Lorenz E, Arbour N, Schwartz DA. Expression and regulation of TLR4 in human bronchial epithelial cells [abstract]. Am J Respir Crit Care Med 2000;161:A151. |
| 27. | Hsu H, Shu HB, Pan MG, Goeddel DV. TRADD-TRAF2 and TRADD-FADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell 1996;84:299–308. |
| 28. | Look DC, Rapp SR, Keller BT, Holtzman MJ. Selective induction of intercellular adhesion molecule-1 by interferon-γ in human airway epithelial cells. Am J Physiol 1992;263:L79–L87. |
| 29. | Bloemen PG, van den Tweel MC, Henricks PA, Engels F, Wagenaar SS, Rutten AA, Nijkamp FP. Expression and modulation of adhesion molecules on human bronchial epithelial cells. Am J Respir Cell Mol Biol 1993;9:586–593. |
| 30. | Look DC, Pelletier MR, Taguchi M, Holtzman. MJ. Adenoviral mechanisms that inhibit intercellular adhesion molecule-1 gene expression in airway epithelial cells [abstract]. Am J Respir Crit Care Med 1997; 155:A460. |
| 31. | Kalvakolanu DV, Bandyopadhyay SK, Harter ML, Sen GC. Inhibition of interferon-inducible gene expression by adenovirus E1A proteins: block in transcriptional complex formation. Proc Natl Acad Sci USA 1991;88:7459–7463. |
| 32. | Datta PK, Bagchi S. Repression of transforming growth factor β1 promoter by the adenovirus oncogene E1A. Identification of a unique GC-rich sequence as a target for E1A repression. J Biol Chem 1994; 269:25392–25399. |
| 33. | Masui T, Wakefield LM, Lechner JF, LaVeck MA, Sporn MB, Harris CC. Type β transforming growth factor is the primary differentiation-inducing serum factor for normal human bronchial epithelial cells. Proc Natl Acad Sci USA 1986;83:2438–2442. |
| 34. | Takizawa H. Airway epithelial cells as regulators of airway inflammation. Int J Mol Med 1998;1:367–378. |
| 35. | Roberts CR. Is asthma a fibrotic disease? Chest 1995;107:S111–S118. |
| 36. | Keatings VM, Collins PD, Scott DM, Barnes PJ. Differences in interleukin-8 and tumor necrosis factor-α in induced sputum from patients with chronic obstructive pulmonary disease or asthma. Am J Respir Crit Care Med 1996;153:530–534. |
| 37. | Thompson AB, Daughton D, Robbins RA, Ghafouri MA, Oehlerking M, Rennard SI. Intraluminal airway inflammation in chronic bronchitis. Am Rev Respir Dis 1989;140:1527–1537. |
