JVI Figure table search 04
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
JVI.00721-06v1
80/22/11241    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tamanini, A.
Right arrow Articles by Cabrini, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tamanini, A.
Right arrow Articles by Cabrini, G.

 Previous Article  |  Next Article 

Journal of Virology, November 2006, p. 11241-11254, Vol. 80, No. 22
0022-538X/06/$08.00+0     doi:10.1128/JVI.00721-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Interaction of Adenovirus Type 5 Fiber with the Coxsackievirus and Adenovirus Receptor Activates Inflammatory Response in Human Respiratory Cells{triangledown}

Anna Tamanini, Elena Nicolis, Alberto Bonizzato, Valentino Bezzerri, Paola Melotti, Baroukh M. Assael, and Giulio Cabrini*

Laboratory of Molecular Pathology, Cystic Fibrosis Center, University-Hospital of Verona, Piazzale Stefani 1, 37126 Verona, Italy

Received 10 April 2006/ Accepted 24 August 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The innate immune response to adenovirus (Ad)-derived gene transfer vectors has been shown to initiate immediately after interaction of Ad with respiratory epithelial cells, through the induction of extracellular signal-regulated kinase 1 and 2 (ERK1/2) and JNK mitogen-activated protein kinase (MAPK), nuclear factor {kappa}B (NF-{kappa}B), and different proinflammatory genes. Ad serotypes 2 or 5 (Ad2/5) enter respiratory epithelia after initial binding of fiber with the coxsackievirus-adenovirus receptor (CAR) or, alternatively, with cell surface heparan sulfate glycosaminoglycans. Ad2/5 internalization is triggered by binding of penton base to cellular RGD-binding integrins. Here we investigated the role of the Ad5 surface domain proteins constituting the vector capsid, namely, the fiber, the penton base, and the hexon, on the transmembrane signals leading to the transcription of the different proinflammatory genes in the human respiratory A549 cell line. Interaction of Ad fiber with CAR activates both ERK1/2 and JNK MAPK and the nuclear translocation of NF-{kappa}B, whereas no activation was observed after exposing A549 cells to penton base and hexon proteins. Moreover, interaction of Ad fiber with CAR, but not heparan sulfate proteoglycans, promotes transcription of the chemokines interleukin-8, GRO-{alpha}, GRO-{gamma}, RANTES, and interferon-inducible protein 10. These results identify the binding of Ad5 fiber with the cellular CAR as a key proinflammatory activation event in epithelial respiratory cells that is independent of the transcription of Ad5 genes.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Replication-defective adenoviruses (Ad) belonging to subgroup C, serotypes 2 (Ad2) and 5 (Ad5), have been extensively studied for the transfer of therapeutic genes into different organs, aimed to the cure of monogenic diseases like cystic fibrosis or complex disorders such as cancer, since they are easy to produce and purify, allow packaging of large genes, and have a wide tissue tropism (for review, see reference 42). First-generation Ad vectors were derived from wild-type Ad2 or Ad5, in which the early genes of the E1 and E3 regions were deleted to render them replication incompetent and to allow the gene of interest to be inserted (33). The application of first-generation E1-E3-deleted Ad vectors to transfer genes into different animal and human tissues induces a pronounced cytotoxic immune response. Extensive investigations revealed that deletion of the genes of the E1-E3 regions was not sufficient to abolish the residual expression of neo-synthesized viral structural proteins, which resulted in major histocompatibility complex class I-associated presentation of viral peptides to immune effector cells. This led to lysis of the target cells which were successfully transduced with the transgene (45). To circumvent the cytotoxic T-lymphocyte response, second-generation Ad-derived vectors with further deletions in the E2 or E4 regions were produced. Deletion of almost the entire viral coding region, as in the helper-dependent or "gutless" Ad vectors, has proved to abolish expression of residual viral proteins, increase efficiency of gene transfer, prolong duration of transgene expression, and minimize the immune response (16, 25).

However, activation of the innate arm of the immune response by Ad vectors has been observed independently of viral gene expression. Massive induction of chemokines like interferon-inducible protein 10 (IP-10), monocyte chemoattractant protein 1, and macrophage inflammatory protein 2 (MIP-2) was found to intervene within 1 hour in murine liver after systemic administration of a first-generation Ad vector (27) and, thus, before viral gene expression. The concept of capsid-dependent immune activation has been recently strengthened after observing induction of tumor necrosis factor alpha (TNF-{alpha}), RANTES, MIP-1{alpha}, MIP-1ß, MIP-2, and IP-10 upon systemic delivery of a helper-dependent gutless Ad vector in the same animal model (28). Epithelial cells are able to respond directly to vector capsid, as shown by the induction of IP-10 in murine renal epithelial cells (4) and of the intercellular adhesion molecule 1 (ICAM-1) in human respiratory cells in vitro (38). That epithelial cells can play an autonomous role is not surprising, considering that surface epithelial cells of respiratory mucosa are not merely a physical barrier to microorganisms but also play a role in triggering proinflammatory signals soon after pathogen interactions, mainly by driving the recruitment of effector cells, including neutrophils, monocytes/macrophages, and natural killer cells. For this reason, the potential role of surface tracheo-bronchial and alveolar epithelial cells in initiating the innate immune response to gene transfer vectors deserves careful investigation.

Different transmembrane signals are elicited during Ad binding and entry into host cells, some of them being potentially involved in the induction of the early innate response (for review, see reference 20). Initial binding and internalization of Ad2 or Ad5 within host cells are known to activate a variety of kinases, such as p125FAK (focal adhesion kinase), p130CAS (Crk-associated substrate), p85/phosphoinositide-3-OH kinase (PI3K), and protein kinase A (PKA), within 15 to 30 min upon exposure (19, 36). Empty capsid or transcription-defective Ad2/5 activate mitogen-activated protein kinases (MAPK) of the extracellular signal-regulated kinase (ERK), p38, and JNK branches within 15 to 30 min and nuclear factor {kappa}B (NF-{kappa}B) within 2 to 4 h (4, 6, 38, 39). Of relevance to the innate response, activation of MAPK (ERK, p38, and JNK) and NF-{kappa}B has been demonstrated to promote the Ad-dependent early onset induction of interleukin-8 (IL-8), IP-10, RANTES, and ICAM-1 in both renal and respiratory epithelial cells (7, 23, 38, 39).

The Ad surface domains which are responsible for the transmembrane signals activating the early innate response after the interaction of the Ad vector capsid with host cell receptors remain to be fully explored. The surface of the Ad capsid is composed of 240 hexon protein trimers forming the 20 triangular faces of the icosahedron and 12 penton structures located at the 12 vertices. Each penton is composed of a spike-shaped protrusion composed of five penton base subunits and a trimeric fiber which presents a globular projecting extremity, or "knob" region, and a long domain termed the "shaft," presenting 22 residue repeats, homologous in Ad2 and Ad5. The fiber knob, the most peripheral domain of Ad2/5, allows initial binding of Ad2/5-derived vectors to a 42-kDa glycoprotein receptor known as the coxsackievirus and Ad receptor (CAR) (2). Initial attachment is also mediated by binding of Ad2/5 vectors to cell surface heparan sulfate glycosaminoglycans (HS-GAGs) (9, 10), through a KKTK sequence located in the fiber shaft (35). After fiber binding, the RGD sequence of the penton base of Ad2/5 interacts with {alpha}v integrins (44), triggering vector internalization through clathrin-coated vesicles. Ad penetration in the cytoplasm occurs 10 min after internalization, and partially uncoated virion traffics along microtubules and reaches the nuclear pore complex by 30 to 40 min, where binding of hexon to histone H1 facilitates entry of viral DNA into the nucleus (12, 40). Although this multistep process involves the {alpha}v integrin-mediated activation of several kinases, such as p125FAK, p130CAS, PI3K, and PKA, and initial studies have investigated the mechanisms of activation of MAPK during Ad entry (6, 18, 19, 36, 39), little is known on the precise role of the binding of penton base with {alpha}v integrins and of the fiber knob with CAR on the proinflammatory signaling leading to expression of chemokines and cytokines (20). Moreover, no studies have been performed yet on the effect of the interaction of Ad fiber shaft with cellular HS-GAGs. Therefore, different pieces of evidence support that the whole Ad capsid is able to trigger a series of early signals resulting in inflammatory response, but little is known of the specific role of the different Ad surface domains in this process.

Understanding the role of the surface domain of Ad vectors in the early induction of the innate response mediated by epithelial cells will have an impact on the development of future Ad-derived vectors. Here we investigated the role of the main Ad surface proteins on the early transmembrane signals, namely ERK1/2, JNK, and the transcription factor NF-{kappa}B, which are involved in the Ad capsid-dependent proinflammatory response. The main observation indicates that early signaling leading to transcription of several inflammatory genes is dependent on the binding of the Ad fiber knob with CAR.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and virus production. Human A549 alveolar type II-derived epithelial cells were obtained from the American Type Culture Collection (Manassas, VA) and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% low-endotoxin fetal bovine serum (FBS; Bio-Whittaker, Walkersville, MD) at 37°C in a 5% CO2-95% air humidified atmosphere. The E1-E3-deleted Ad5-derived vector carrying the cystic fibrosis gene Ad.CFTR was obtained as stock from Transgene SA (Strasbourg, France) and propagated in 293 human embryonal kidney cells (33). The virus was purified by four freeze-thawing cycles, followed by two successive bandings on CsCl gradients (30). Purified virus was dialyzed against 10 mM Tris-HCl (pH 8) and 1 mM MgCl2, aliquoted with the addition of 10% glycerol (virus suspension buffer), and stored at –80°C until use. The concentration of purified Ad was determined by the absorbance at 260 nm measured according to the method of Mittereder et al. (24), who assumed the conversion factor of 1 optical density unit of Ad corresponding to 1.1 x 1012 virions.

Production and purification of the knob domain of the Ad5 fiber protein in Escherichia coli. A DNA fragment encoding the whole Ad5 fiber knob domain and several flanking amino acids from the fiber shaft (amino acids 387 to 581) was amplified from viral DNA by PCR using specific primers designed to facilitate the insertion of the PCR product directly into the bacterial expression vector, by modifying the protocol originally described by Henry et al. (15). The oligonucleotides used were Ad5K-D (CTC GAA TTC ATG GGT GCC ATT ACA GTA GGA AAC) and Ad5K-R (ACG TGC CTG CAG TTA TTC TTG GGC AAT GTA TGA), which contain EcoRI and PstI restriction sites (underlines), respectively. The PCR product was cloned between the EcoRI and PstI sites of vector pKK223-3 (Amersham-Pharmacia) and transformed into strain JM109 (Stratagene) for expression of the knob protein. To induce knob expression, overnight cultures in LB-ampicillin broth were diluted 100-fold and grown until mid-log phase (optical density at 600 nm of 0.6), at which time they were adjusted to 1.5 mM isopropyl ß-D-thiogalactopyranoside. After shaking at 37°C for 4 h, the bacterial cells were collected by centrifugation. Extraction of the knob domain from JM109 bacterial cells as a soluble protein was performed as previously described with minor modification (15). Briefly, bacterial cells were incubated with lysozyme and nonionic detergent in order to obtain cell lysis. Viscosity due to chromosomal DNA release from broken cells was reduced by DNase I digestion, and the cell wall debris was removed by centrifugation at 13,000 x g for 20 min. The supernatant was then batch absorbed with DEAE-Sephadex equilibrated with 0.2 M NaCl, 50 mM Tris-HCl, pH 8, 1 mM EDTA to remove most of the DNA and a few proteins. Ammonium sulfate precipitation, at 100% saturation, was used to concentrate the material. Precipitated proteins were dissolved in 20 mM Tris-HCl, pH 8, 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride and dialyzed against that buffer at 4°C for 48 h. After dialysis, proteins were centrifuged and the supernatant was adsorbed on a Q-Sepharose FF protein liquid chromatography column (1.6 by 30 cm) which was equilibrated in the same buffer. Development of the column with a 0 to 80 mM NaCl gradient eluted the knob at 35 mM NaCl in a single peak, which was pooled and concentrated using an ultrafiltration cell (Amicon Corp., Lexington, Mass.) with a 3,000-Da cutoff ultrafiltration membrane (Millipore Corporation, Bedford, Mass.). The purified protein was electrophoretically pure as judged by 15% sodium dodecyl sulfate (SDS)-polyacrylamide gels and Coomassie blue staining, as shown in Fig. 1B.


Figure 1
View larger version (73K):
[in this window]
[in a new window]
  		FIG. 1. Purification of viral capsid proteins. (A) Separation of viral proteins on 10% SDS-PAGE. Ten to twenty µg of each purified viral protein was loaded. Lane 1, hexon (Hx; 130 kDa); lane 2, full-length fiber incubated in PAGE sample buffer at 94°C for 3 min, resulting in monomeric fiber (mF; 65 kDa); lane 3, full-length fiber incubated in PAGE sample buffer at room temperature to maintain the trimeric form (tF; 180 kDa); lane 4, penton base (Pb; 85 kDa). The molecular mass markers are indicated on the right. (B) Separation of recombinant knob on 15% SDS-PAGE. The knob domain as trimeric form (lane 1, tK; 75 kDa) and as monomeric form (lane 2, mK; 25 kDa) was obtained by incubating the recombinant fiber knob at room temperature or at 94°C for 3 min, respectively. Ten µg of purified knob was loaded in each lane. The molecular mass markers are indicated on the left. (C to E) A cell infection assay was performed as reported in Material and Methods by incubating A549 cells with buffer (C), purified fiber (1 µg/ml) (D), or Ad5 (MOI, 10) (E).

 
Purification of soluble Ad fiber, hexon, and penton base. We performed the purification of soluble Ad fiber, hexon, and penton base by using as starting material the top of the CsCl gradient obtained during Ad5 purification. The purification process of this part of the gradient, which contains nonincorporated viral proteins, required multistep procedures, including neutral salt precipitation, ion exchange, adsorption, and affinity chromatography, as previously described with minor modifications (5). In brief, viral proteins were precipitated with ammonium sulfate, extensively dialyzed against 50 mM Na-phosphate, pH 6.8, centrifuged at 100,000 x g at 4°C for 2 h to remove possible incomplete low-density virus particles, and initially fractionated by chromatography on DEAE-Sephadex A-50 (Sigma, St. Louis, MO) in the same starting buffer in which a stepwise gradient from 0 to 150 and 400 mM NaCl was applied. The flowthrough volume (0 mM NaCl) contained the fiber as major protein together with other lower-molecular-weight proteins. Penton, with core protein V, was eluted at 150 mM NaCl, and hexon eluted at 400 mM NaCl. Final purification of the fiber was obtained after gel filtration with Sephacryl S-200 high-resolution resin (Amersham Biosciences, Uppsala, Sweden) in a 1.6- by 30-cm column with 50 mM Na-phosphate, 150 mM NaCl, pH 6.8, and the major peak eluted was analyzed by SDS-polyacrylamide gel electrophoresis (PAGE). Protein solutions were solubilized in SDS-PAGE sample buffer (62 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 5% ß-mercaptoethanol, 0.2% bromophenol blue) for 3 min at 94°C prior to PAGE. A single band of the apparent molecular mass of 180 kDa, corresponding to the trimeric full-length fiber, was observed by adding PAGE sample buffer at room temperature, as shown in Fig. 1A, lane 3. The same sample incubated at 94°C for 3 min gave a single band of the apparent molecular mass of 65 kDa, corresponding to the monomeric form of the full-length fiber, as shown in Fig. 1A, lane 2. Penton base was purified from penton and core protein V, eluted via DEAE-Sephadex A-50 chromatography, firstly by retention of the core protein V onto a hydroxyapatite 1- by 10-cm column (Bio-Gel HTP; Bio-Rad Laboratories, Richmond, CA) equilibrated with 10 mM K-phosphate, pH 6.8, and subsequently by subtracting fiber not covalently attached to the penton base, from the whole penton using affinity chromatography with a polyclonal antibody raised against the Ad fiber (Alfa Biotech, Pomezia, Italy) covalently immobilized with dimethyl-pimelidate on Sepharose-protein A beads (Amersham Biosciences, Uppsala, Sweden). The final purification product run on SDS-PAGE gave the expected apparent mass for penton base of 85 kDa, as shown in Fig. 1A, lane 4. Hexon was directly utilized after dialysis of the 400 mM NaCl fraction from the DEAE-Sephadex A-50 chromatography. SDS-PAGE confirmed a single band of the expected molecular mass of 130 kDa (Fig. 1A, lane 1).

Cell infection assay. In order to exclude the presence of contaminating intact particles in the fiber preparation, a cell infection assay was performed as previously described (9, 10), according to the methods of Wickham et al. (44). Briefly, A549 cells were seeded onto chamber slides, in complete DMEM supplemented with 10% FBS and left to reach confluence, to obtain a standard surface area. Ad5 (multiplicity of infection [MOI], 10) or fiber (1 µg/ml) was added. Wells were washed twice with phosphate-buffered saline (PBS) before overnight incubation with tissue culture medium. Cells were fixed with acetone, blocked with 10% pig serum in PBS for 30 min, and incubated with a primary antibody directed against Ad hexon protein (Chemicon) or mouse immunoglobulins as a negative control at 10 µg/ml in 10% pig serum-PBS for 1 h. After washing, cells were incubated for 30 min with a 1:100 dilution of fluorescein isothiocyanate-conjugated rabbit anti-mouse antibody (Sigma) in 1% Evans blue in PBS. As shown in Fig. 1E, the fluorescent signal of the viral proteins in the nucleus after synthesis in the cytoplasm was very evident. Specificity of the signal was checked with an irrelevant murine antibody (not shown). The incubation with fiber (1 µg/ml) or virus-free buffer did not induce fluorescent nuclear signal, as shown in the sample microscopic fields in Fig. 1C and D. The presence of nuclear neo-synthesized viral protein after incubation with purified fiber was completely excluded in all slides, demonstrating that purified fiber was free of intact adenovirus particles.

Production and purification of sCAR-D1. The soluble extracellular N-terminal domain 1 (D1) of the human coxsackievirus-adenovirus receptor (sCAR-D1) was expressed and purified as previously described (9), according to the methods of Freimuth and Coll (11).

Lipopolysaccharide (LPS) measurement. The levels of endotoxin present in the CAR-D1 and knob protein solutions were determined using lyophilized Limulus amebocyte lysate with the Pyrogent Plus single test kit, 0.06 EU/ml, and Pyrogent Plus single test kit, 0.125 EU/ml (Cambrex Bio Science, Walkersville, MD) according to the manufacturer's instructions. Knob and sCAR-D1 protein preparations were tested at 1 µg/ml and 200 µg/ml concentrations. The endotoxin concentrations measured in the two protein solutions were within a range corresponding to 6 to 12 pg/ml.

Exposure of the cells to whole Ad and to different viral proteins. A549 cells grown on 60-mm-diameter dishes (1.5 x 106 cells/dish) were starved for 18 h in serum-free DMEM (1 ml) to avoid activation of MAPK by the growth factors contained in the serum. After starvation, A549 cells were exposed to Ad.CFTR virus suspension (10 to 20 µl) at the specified MOI or to the knob domain of Ad5 (1 µg/ml), full-length fiber (1 µg/ml), penton base (1 µg/ml), or hexon (1 µg/ml). In selected experiments, Ad.CFTR was preincubated with either sCAR-D1 (200 µg/ml) or heparin (0.1, 1, or 10 µg/ml).

Preparation of cell lysates. Cells were incubated as indicated above and then washed three times with ice-cold PBS, scraped, and collected by centrifugation at 13,000 x g for 5 min at 4°C. The pellet was suspended in 50 µl MAPK lysis buffer containing 20 mM Tris-HCl (pH 7.4), 137 mM NaCl, 2 mM EDTA, 10% glycerol, 0.1% SDS, 0.5% deoxycholic acid, 1% Triton X-100, 25 mM {alpha}-glycerol phosphate, 1 mM Na3VO4, 20 µM leupeptin, 0.2 mM phenylmethylsulfonyl fluoride, and 25 µg/ml aprotinin and incubated at 4°C for 30 min. Cell lysate was cleared by centrifugation at 13,000 x g at 4°C for 20 min. Protein concentration was determined by the method of Lowry et al. (21) after precipitation with 5% trichloroacetic acid, utilizing bovine serum albumin as a standard. Aliquots were stored at –80°C.

ERK activity assay. The activity of ERK1/2 was determined with the Biotrak MAP kinase assay (Amersham Pharmacia Biotech) as previously described (38), with minor modifications. Briefly, cell lysates (10 µg protein/15-µl aliquot) were incubated at 30°C for 30 min with 10 µl of a synthetic peptide substrate containing a sequence of the epidermal growth factor receptor (EGFR) with a single phosphorylation site for ERK1/2. The reaction was started by adding 1 µCi/5 µl of Mg-[{gamma}-32P]ATP (Amersham) and stopped with 10 µl of stop reagent. Samples were spotted on 3-cm-diameter disks of binding paper. After washing twice in 1% acetic acid and twice in water, disks were placed in scintillation vials, and radioactivity was counted in a Packard Tri-Carb 2100 TR liquid scintillation analyzer. Counts from the endogenous protein phosphorylation and nonspecific binding of Mg-[{gamma}-32P]ATP to filter were subtracted.

Immunoprecipitation and Western blot analysis of JNK1. Analysis of JNK1 was performed as previously described (38), with minor modifications. Briefly, aliquots of cell lysates (50 µg protein per sample) were incubated at 4°C for 2 h with 10 µg/ml anti-JNK polyclonal antibody (Santa Cruz Biotechnology) in 100 µl MAPK lysis buffer as described above. Samples were then incubated overnight at 4°C with 10 µl protein A-Sepharose beads washed three times with MAPK lysis buffer and two times with kinase buffer (20 mM morpholinepropanesulfonic acid, pH 7.2, 25 mM ß-glycerol phosphate, 5 mM EGTA, 1 mM dithiothreitol, 1 mM Na3VO4) and then incubated at 30°C for 30 min in 30 µl containing 1 µg of c-Jun(1-169)-glutathione-S-transferase (GST) as substrate (Upstate Biotechnology), 5 µCi Mg-[{gamma}-32P]ATP, 10 µM ATP, in the same kinase buffer. Phosphorylation reactions were terminated by dilution with 5x SDS-PAGE sample buffer. Samples were boiled for 3 min at 95°C and clarified by centrifugation, and 32P-labeled c-Jun (1-169)-GST products were separated on 10% SDS-polyacrylamide gels and subjected to autoradiography. Phosphorylation bands on X-ray films were visualized with a Hamamatsu C2400-97 charge-coupled-device (CCD) video camera (Hamamatsu City, Japan). JNK1 protein levels were measured from the same lysates used for the kinase assay. Equal amounts of protein of cell lysates were separated by SDS-PAGE and blotted to nitrocellulose, then filters were incubated with a 1:100-diluted polyclonal antibody specific for JNK1 (Santa Cruz Biotechnology) at room temperature for 1 h, and after three washes membranes were incubated with horseradish peroxidase-coupled anti-rabbit immunoglobulin G (Amersham) at a 1:5,000 dilution at room temperature for 1 h and then washed three times. The signal was developed by enhanced chemiluminescence (ECL kit; Amersham Pharmacia Biotech).

Immunofluorescent staining of NF-{kappa}B. Analysis of NF-{kappa}B was performed as previously described (38). Briefly, A549 cells were seeded on eight-well chamber slides (Nunc, Naperville, IL) coated with 10 µg/ml purified collagen (Vitrogen-100; Collagen Corporation, Palo Alto, CA) for 2 h and cultured in complete DMEM supplemented with 10% FBS. Cells were then starved for 18 h in serum-free DMEM before exposure to viral domains or the transcription-defective Ad.CFTR vector in a volume of 5 to 10 µl suspension buffer, as described in Results, and left in place for an additional 4 h. Cells were washed three times with PBS and fixed with 4% paraformaldehyde (wt/vol) in PBS for 20 min at room temperature. After three washes with PBS, cells were permeabilized with methanol at –20°C for 5 min and then air dried for 1 h. Slides were incubated with 5% bovine serum albumin (BSA) in PBS for 90 min at room temperature and then subjected to three 1-h incubations at room temperature with 7 µg/ml rabbit polyclonal antibody against the NF-{kappa}B p65 subunit (sc-109; Santa Cruz Biotechnology Inc., Santa Cruz, CA) in 5% BSA, with a 1:200 dilution of biotinylated goat anti-rabbit immunoglobulin G (Santa Cruz Biotechnology) in 1% BSA-PBS-0.1% Tween 20 and a 1:60 dilution of fluorescein isothiocyanate-conjugated streptavidin (Sigma) in 1% BSA-PBS-0.1% Tween 20. Coverslips were mounted with Prolong antifade (Molecular Probes, Eugene, OR) and stored at room temperature. Fluorescence was examined with a Nikon TMD fluorescence microscope.

Quantitation of transcripts of inflammatory genes. Total RNA from A549 cells was isolated using the High Pure RNA isolation kit (Roche, Mannheim, Germany). Total RNA (2.5 µg) was reverse transcribed to cDNA using the High Capacity cDNA Archive kit and random primers (Applied Biosystems) in a 100-µl reaction mixture. The cDNA (2 µl) was then amplified for 50 PCR cycles using the Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) in an ABI Prism 5700 sequence detection system (Applied Biosystems). The real-time PCRs were performed in duplicates for both target and normalizer genes. Primer sequences and concentrations are shown in Table 1. Primer sets were purchased from Sigma-Genosys (The Woodlands, TX). Results were collected with Sequence Detection software (version 1.3; Applied Biosystems). Relative quantification of gene expression was performed using the comparative threshold (CT) method as described by the manufacturer (Applied Biosystems User Bulletin 2). Changes in mRNA expression level were calculated following normalization to the calibrator gene. The ratios obtained following normalization are expressed as the fold change over nontreated samples.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Primers for quantitation of inflammatory genes

 

   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Full-length fiber and the fiber knob activate ERK1/2. Helper-dependent gutless Ad vectors induce the release of proinflammatory cytokines and chemokines in murine liver within 1 h of systemic delivery (28). In addition, our previous work suggested a direct role of the Ad capsid, since we found that transcription-defective Ad vectors upmodulate the adhesion molecule ICAM-1 in lung epithelial cells, through the activation of the MAPK ERK1/2 and JNK within 30 min, and activation of NF-{kappa}B within 4 h (38). To investigate the role of the major capsid surface proteins in this process, human respiratory epithelial A549 cells were exposed to Ad5 full-length fiber, fiber knob, hexon, and penton base, and MAPK ERK1/2 was assayed by phosphorylation of a peptide substrate. ERK1/2 was activated after exposing A549 cells to full-length fiber or fiber knob, with a peak of maximal activity from 5 to 20 min as shown in Fig. 3A and B, below. Conversely, Ad penton base, which is the most abundant surface protein of the Ad capsid and is known to bind {alpha}v integrins, and hexon did not elicit ERK1/2 substrate phosphorylation (see Fig. 3C and D, below). The globular region of the Ad2/5 fiber, termed the knob, is known to bind the first extracellular domain of the CAR (32). In order to understand whether ERK1/2 phosphorylation was induced by the specific binding of Ad fiber to CAR, the soluble first extracellular domain of CAR (sCAR-D1) was preincubated with either full-length fiber or fiber knob before exposure to cells, as described in Materials and Methods. The potential toxicity of sCAR-D1 in the proinflammatory signals was excluded by testing the absence of effect of sCAR-D1 on TNF-{alpha}-dependent activation of ERK1/2, as shown in Fig. 2B. As shown in Fig. 3A and B, preincubation with sCAR-D1 completely abolished the phosphorylation of ERK1/2 induced by full-length fiber and fiber knob, suggesting that ERK1/2 signaling is dependent on the interaction of the fiber knob with CAR.


Figure 3
View larger version (18K):
[in this window]
[in a new window]
  		FIG. 3. Ad fiber induces ERK1/2 kinase activity. The ERK1/2-dependent phosphorylation of a specific peptide substrate containing a sequence of EGFR was measured in A549 cell lysates. A. Cells exposed to full-length fiber (1 µg/ml; filled circles) or full-length fiber preincubated with sCAR-D1 for 60 min (1 µg/ml and 200 µg/ml, respectively; open triangles). B. Cells exposed to the fiber knob domain (1 µg/ml; filled circles) alone or preincubated with sCAR-D1 (1 µg/ml and 200 µg/ml, respectively; open triangles). C. Cells exposed to penton base (1 µg/ml; filled circles) or an equivalent volume of buffer (open triangles). D. Cells exposed to hexon (1 µg/ml; filled circles) or an equivalent volume of buffer (open triangles). Radioactivity incorporated at each time is expressed as a percentage of the counts obtained at time zero. Data are means ± standard errors of the means of three independent experiments.

 

Figure 2
View larger version (35K):
[in this window]
[in a new window]
  		FIG. 2. Effects of sCAR-D1 on TNF-{alpha}-dependent activation of MAP kinases and IL-8 mRNA. A549 cells were incubated with TNF-{alpha} (50 ng/ml) in the presence or absence of sCAR-D1 (200 µg/ml). (A) JNK activity (30 min); (B) ERK1/2 activity (30 min; mean ± standard error of the mean); (C) IL-8 mRNA (18 h; mean ± standard error of the mean).

 
Activation of JNK1 phosphorylation by the fiber knob. MAPK of the JNK/SAPK pathway has been also shown to be involved in the upregulation of proinflammatory genes by exposing A549 respiratory cells to transcription-defective Ad (38). Therefore, the phosphorylating activity of JNK1 was assayed on c-Jun(1-169)-GST substrate after addition of fiber knob, hexon, or penton base to A549 cells. Ad fiber knob stimulated phosphorylation of c-Jun-GST substrate within 5 to 20 min of exposure, returning to basal levels after 60 min (Fig. 4A). The total amount of JNK1 protein immunoprecipitated was very similar at each time point, as checked by Western blot analysis. As observed for ERK1/2, fiber knob preincubated with the soluble extracellular domain 1 of CAR before exposure to cells abolished the activation effect of fiber knob (Fig. 4B). Human TNF-{alpha} was utilized as a positive control to confirm the integrity of the JNK/SAPK signaling pathway in the same experiment and to exclude the untoward toxicity of sCAR-D1 (as shown in Fig. 2A). Cells incubated with penton base or hexon did not show activation of JNK1 (Fig. 4C and D).


Figure 4
View larger version (44K):
[in this window]
[in a new window]
  		FIG. 4. Ad fiber induces JNK1 kinase activity. Cell lysates were analyzed for JNK1 activity by an immunocomplex assay using c-Jun(1-169)-GST as phosphorylation substrate. Autoradiograms of phosphorylated substrate (c-jun-GST) and Western blots of total JNK1 (tJNK1) are shown. Cells were exposed to fiber knob (1 µg/ml) (A), fiber knob (1 µg/ml) preincubated with sCAR-D1 (200 µg/ml) (B), penton base (1 µg/ml) (C), or hexon (1 µg/ml) (D). TNF-{alpha} (50 ng/ml) was used as a positive control. Results are representative of three independent experiments.

 
Full-length fiber and fiber knob, but not hexon or penton base, activate NF-{kappa}B. Activation of the transcription factor NF-{kappa}B plays a key role in the expression of a large series of proinflammatory genes. As previously reported, transcription-defective Ad is able to activate NF-{kappa}B after an exposure time ranging from 4 to 24 h (23, 38). NF-{kappa}B activation was assayed by detecting the nuclear translocation of the p65 subunit of NF-{kappa}B by immunofluorescence. NF-{kappa}B translocation was studied 4 h after exposure of A549 cells to full-length fiber, fiber knob, hexon, or penton base. As shown in Fig. 4A and C, full-length fiber and fiber knob activated nuclear translocation of NF-{kappa}B, in association with a decrease of the cytoplasmic fluorescent signal. In contrast, the signal of p65 remained in the perinuclear region when cells were exposed to full-length fiber or fiber knob preincubated with sCAR-D1 (Fig. 5B and D). Again, hexon, penton base, sCAR-D1 alone, or suspension buffer were unable to induce nuclear translocation of NF-{kappa}B (Fig. 5E to H). The toxicity of sCAR-D1 on activation of NF-{kappa}B was excluded, as 200 µg/ml sCAR-D1 did not inhibit TNF-{alpha}-dependent signal, as shown in Fig. 6E and F. Moreover, LPS induced NF-{kappa}B activation only at a very high concentration (10 µg/ml), which was 6 orders of magnitude higher than that found as the tracer amount in knob and sCAR-D1 preparations (Fig. 6D).


Figure 5
View larger version (128K):
[in this window]
[in a new window]
  		FIG. 5. Fiber knob induces the nuclear translocation of NF-{kappa}B. A549 cells were incubated for 4 h with full-length fiber (1 µg/ml) (A), full-length fiber (1 µg/ml) preincubated with sCAR-D1 (200 µg/ml) (B), fiber knob (1 µg/ml) (C), fiber knob (1 µg/ml) preincubated with sCAR-D1 (200 µg/ml) (D), hexon (1 µg/ml) (E), penton base (1 µg/ml) (F), suspension buffer (G), or sCAR-D1 alone (200 µg/ml) (H).

 

Figure 6
View larger version (143K):
[in this window]
[in a new window]
  		FIG. 6. Effects of LPS and TNF-{alpha} on NF-{kappa}B translocation. A549 cells were incubated for 4 h with LPS, TNF-{alpha} with or without sCAR-D1, or buffer alone, as indicated. Cell autofluorescence (immunofluorescence with irrelevant rabbit immunoglobulin instead of primary antibody against the p65 subunit of NF-{kappa}B) (A), buffer alone (B), LPS (0.1 µg/ml) (C), LPS (10 µg/ml) (D), TNF-{alpha} (50 ng/ml) (E), or TNF-{alpha} (50 ng/ml) plus sCAR-D1 (200 µg/ml) (F).

 
ERK1/2 signaling is independent of interaction with HS-GAGs. We previously demonstrated that Ad2/5 utilizes HS-GAGs as receptors for binding and infection of target cells (10), independently of the expression of CAR (9). Recently, the 91KKTK94 sequence located in the third pseudorepeat of the fiber shaft has been demonstrated to be the HS-GAG binding site (35). Since full-length fiber includes both the 91KKTK94 sequence of the shaft and the globular knob region involved in binding to CAR, we investigated the potential role of HS-GAGs in the activation of ERK1/2. We preincubated full-length fiber with the soluble HS analogue heparin before exposure to A549 cells (9). Activation of ERK1/2 even after competitive inhibition of the HS-binding site of the fiber shaft with heparin, as illustrated in Fig. 7C, indicates that ERK1/2 signaling intervenes independently of the binding of the fiber shaft with cell surface HS-GAGs and strengthens the role of the interaction of the fiber knob with CAR in inducing this MAP kinase. The results presented so far from experiments which utilize separately the major Ad surface proteins have been verified with Ad.CFTR, a replication-defective Ad vector. Ad.CFTR preincubated with the HS soluble analogue heparin still induces activation of ERK1/2, as shown in Fig. 7B, confirming the irrelevance of cell surface HS-GAGs in inducing this pathway. When Ad.CFTR was preincubated with sCAR-D1, a condition under which Ad2/5 is able to accomplish full infection through HS-GAGs (9), the activation of ERK1/2 was completely abolished, as shown in Fig. 7A, confirming the key role of CAR in the activation of ERK1/2 signaling after the initial binding of Ad capsid to target cells.


Figure 7
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 7. Effect of heparin on induction of ERK1/2 kinase. The ERK1/2-dependent phosphorylation of a specific peptide substrate containing a sequence of EGFR was measured in A549 cell lysates obtained after exposure to Ad.CFTR (MOI of 25; filled circles) or Ad.CFTR preincubated with sCAR-D1 (MOI of 25 and 200 µg/ml, respectively; open triangles) (A), Ad.CFTR (MOI of 25; filled circles) or to Ad.CFTR preincubated with heparin (MOI of 25 and 1 µg/ml, respectively; open triangles) (B), or full-length fiber (1 µg/ml; filled circles) or full-length fiber preincubated with heparin (1 and 10 µg/ml, respectively; open triangles) (C). Data are means ± standard errors of the means of three independent experiments.

 
Nuclear translocation of NF-{kappa}B is independent of Ad binding to HS-GAGs. Since HS-GAGs could in principle activate proinflammatory signals through activation of NF-{kappa}B, also independently of the MAPK pathway, we investigated the induction of NF-{kappa}B by using immunofluorescence. We exposed A549 cells to full-length fiber or the whole Ad.CFTR vector preincubated with the HS soluble analogue heparin. Both full-length fiber and Ad.CFTR in which HS binding sites were competitively inhibited with heparin induced activation of NF-{kappa}B, as shown in Fig. 8A and B, excluding a role of cell surface HS-GAGs. Conversely, NF-{kappa}B activation was abolished when the interaction between the whole Ad.CFTR vector and CAR was competitively inhibited by preincubation with sCAR-D1, as shown in Fig. 8C and D. This confirms the direct involvement of Ad fiber-CAR binding in the activation of NF-{kappa}B when Ad virus interacts with host cells.


Figure 8
View larger version (144K):
[in this window]
[in a new window]
  		FIG. 8. Effect of heparin on induction of NF-{kappa}B. A549 cells were incubated for 4 h with full-length fiber preincubated with heparin (1 and 10 µg/ml, respectively) (A), Ad.CFTR preincubated with heparin (MOI of 25 and 10 µg/ml, respectively) (B), Ad.CFTR preincubated with sCAR-D1 (MOI of 25 and 200 µg/ml, respectively) (C), or Ad.CFTR preincubated with heparin and sCAR-D1 (MOI of 25, 10 µg/ml, and 200 µg/ml, respectively) (D). Slides were incubated with a primary antibody directed against the NF-{kappa}B p65 subunit. Results are representative of three independent experiments.

 
Ad5 fiber induces transcription of inflammatory genes. Taken together, the results presented here suggest that the binding of Ad fiber to CAR has a key role in the early induction of the intracellular pathways ERK1/2, JNK, and NF-{kappa}B. In order to ascertain whether these early signals are sufficient to induce transcription of proinflammatory genes, we exposed A549 cells to the replication-defective Ad.CFTR vector and Ad5 full-length fiber to measure gene expression by quantitative reverse transcription-PCR. A panel of genes involved in the innate immune response was chosen, with special regard to those induced by Ad-derived vectors (20), wild-type Ads, and other respiratory viruses, the most representative being IL-8, a potent neutrophil chemoattractant (7). Both the replication-defective vector Ad.CFTR and full-length fiber markedly induced IL-8 mRNA, as shown in Fig. 9. To understand whether IL-8 induction was due to the binding of full-length fiber to CAR or to HS-GAGs, full-length fiber was preincubated with either sCAR-D1 or heparin to inhibit the binding to CAR or HS-GAG receptors, respectively. Only preincubation with sCAR-D1 completely inhibited IL-8 mRNA induction, indicating that the interaction of fiber with CAR triggers a series of transmembrane signals eventually leading to the transcription of the IL-8 gene. The potential toxicity of sCAR-D1 on IL-8 transcription was excluded by checking the lack of effect on TNF-{alpha}-dependent induction, as shown in Fig. 2C. A dose response of LPS is shown in Fig. 9C. Importantly, a small increase of IL-8 mRNA was found at LPS concentrations starting from 0.1 µg/ml, while induction comparable to that obtained with fiber was observed only with LPS at 10 µg/ml, a concentration 6 orders of magnitude higher than that found in the sCAR-D1 preparation. To generate a more complete picture of the effect of CAR stimulation on the inflammatory response, we extended the quantitation of transcripts to genes encoding chemokines attracting neutrophils or mononuclear immune cells into inflammatory sites (GRO-{alpha}, GRO-{gamma}, MIP-1{alpha}, RANTES, and IP-10) or proinflammatory or regulatory cytokines (TNF-{alpha}, IL-6, IL-1ß, and IL-10). Interestingly, activation of CAR via binding of Ad fiber clearly induced the chemokines GRO-{alpha}, GRO-{gamma}, RANTES, and IP-10, besides IL-8, whereas no induction was observed on the proinflammatory and regulatory genes included in our panel (Fig. 9B).


Figure 9
View larger version (20K):
[in this window]
[in a new window]
  		FIG. 9. Full-length fiber induces transcription of inflammatory genes. A. A549 cells (2 x 105 cells) were grown in 2-cm-diameter wells and incubated with Ad.CFTR vector (MOI, 100), full-length fiber (1 and 0.1 µg/ml), full-length fiber (1 or 0.1 µg/ml) preincubated for 1 h at 37°C with sCAR-D1 (200 and 20 µg/ml), sCAR-D1 alone (200 µg/ml), full-length fiber (1 µg/ml) preincubated for 1 h at 37°C with heparin (10 or 1 or 0.1 µg/ml), or heparin alone (10, 1, or 0.1 µg/ml) for 18 h before RNA extraction, reverse transcription, and IL-8 quantitation, as described in Materials and Methods. Values are means ± standard errors of the means of four representative experiments. B. Quantitation of the indicated inflammatory genes was performed as for IL-8 by incubated full-length fiber (1 µg/ml) in the absence (open symbols) or presence (close symbols) of sCAR-D1 (200 µg/ml). Results are representative of three independent experiments. C. IL-8 induction by LPS. A549 cells were incubated for 18 h with the indicated concentrations of LPS, and IL-8 mRNA was quantified as specified above.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Application of Ad-derived vectors for gene therapy has been largely limited by the induction of the innate and adaptive arms of the immune response against the same cells in which the gene of interest has been transferred, as extensively studied in different disease models, including the transfer of the cystic fibrosis gene into surface respiratory epithelial cells. Multiple efforts have been made to increase the efficiency and safety of Ad vectors by abolishing the antiviral immune response, mainly by preparing a new generation of Ad vectors with a complete absence of de novo viral protein synthesis, such as the helper-dependent gutless Ad vectors (16). However, the early onset of transmembrane and intracellular signals, such as MAPK and NF-{kappa}B, and expression of proinflammatory genes was observed even after application of transcription-inactivated and gutless Ad vectors (28, 37). Thus, questions arise as to whether the simple interaction of the surface protein domains constituting the Ad capsid with the corresponding cell receptors is able per se to activate early proinflammatory signals. We show here that the interaction of the Ad fiber knob with the cellular receptor CAR activates very rapidly MAPK ERK1/2 and JNK together with the transcription factor NF-{kappa}B, which are involved in the induction of transcription of the chemokines IL-8, GRO-{alpha}, GRO-{gamma}, RANTES, and IP-10, by exposing respiratory A549 cells to the major Ad capsid protein components, in a model system which is completely independent of de novo expression of any Ad gene. Therefore, our data (i) confirm that the Ad5 capsid components are sufficient per se to induce early proinflammatory signals in respiratory epithelial cells, (ii) identify the interaction between Ad5 fiber and CAR as being responsible for the induction of MAPKs, NF-{kappa}B, and gene transcription of different chemokines (IL-8, GRO-{alpha}, GRO-{gamma}, RANTES, and IP-10) and (iii) make unlikely the participation of the other major Ad5 surface domains, such as hexon and penton base, in triggering the Ad5-dependent early proinflammatory events.

The timing of Ad2/5 entry in A549 cells was elegantly elucidated by Greber and Coworkers (12, 40). Fiber dissociates from Ad capsid at 5 min after binding to primary receptors, now identified as CAR and/or HS-GAGs (2, 9, 10), and coated-pit-mediated internalization initiates 10 min later by interaction of the penton base with RGD-binding {alpha}v integrins (44). About 20 to 30 min after initial binding, acidification of the endosome triggers penetration of the partially dismantled capsid into the cytoplasm, and then capsid reaches the nuclear pore along the microtubule network in 35 to 45 min. Contact of the capsid hexon with histone H1 allows import of the Ad DNA within the nucleus at least 45 min post-initial binding to the primary receptors (12, 40). Thus, the time of activation of MAPK ERK1/2 and JNK by full-length fiber and fiber knob, which we observed to happen from 5 to 20 min postexposure, is compatible with signaling starting just after binding of the fiber knob to CAR. In our experimental model, we estimate that the concentration of free fiber able to activate a proinflammatory signal is higher than that corresponding to the fiber constituting the intact virus. Thus, the multiple interactions of fiber knob with CAR in intact virus may increase the efficiency of the proinflammatory signaling with respect to that elicited by free fiber protein, possibly by clustering different CAR subunits in specific plasma membrane microdomains. That viral capsids are sufficient per se in activating a proinflammatory response could have been inferred by the response elicited by gutless Ad-derived vectors. Importantly, this has been directly reported for other viruses, such as human cytomegalovirus, in which the participation of receptors classically involved in the innate response, such as CD14 and Toll-like receptor 2, have been elegantly demonstrated (8). Interestingly, CAR was first identified as a receptor allowing the attachment of coxsackievirus and Ad virus, whereas its biological functions are still largely unknown. While the colocalization of CAR within the tight junction prompted the speculation of its potential role in maintaining epithelial integrity (43), our original observation that binding of Ad5 fiber knob to CAR activates MAP kinases and NF-{kappa}B suggests auxiliary roles, such as its participation in cell proliferation and innate mechanisms of defense against pathogens. The role of CAR on the activation of ERK1/2 has been previously studied with the fiberless Ad vector AdL.F, which was shown to induce a partial activation of ERK1/2 and IP-10 mRNA compared to a AdCMVßgal vector that had the same surface domains as Ad2/5 and Ad.CFTR (39). In contrast, here we did not observe any residual activation when the interaction of Ad fiber to CAR was blocked with sCAR-D1. This apparent discrepancy can be explained by considering the different cell types studied (renal epithelial versus alveolar type II-derived cells) and the very high concentration of the AdF.L vector (39).

We also observed that the other major Ad surface domains are not involved in MAPK and NF-{kappa}B proinflammatory signaling. Binding of penton base to {alpha}v integrins was shown to trigger Ad2/5 internalization and to activate p125FAK focal adhesion kinase, p130CAS, PI3K, and PKA (18, 19, 44). RGD-binding integrins play different roles in lung inflammation models, such as that of protection from fibrosis upon binding of transforming growth factor ß1 (26), of removing eosinophilic apoptotic bodies (34), and more recently of enhancing IL-1ß-induced release of eotaxin, RANTES, and granulocyte-macrophage colony-stimulating factor upon binding to extracellular matrix components (29). Studies have been performed on the effect of penton base-{alpha}v integrin interactions on the activation of MAPK and NF-{kappa}B, together with the expression of RANTES and IP-10, in HeLa, REC kidney, and P815 mastocytoma cells (6, 39). The present consensus, which indicates that induction of proinflammatory signals is mainly RGD integrin independent (20), is in agreement with the results presented here in A549 respiratory cells. Hexon, the most abundant Ad surface protein, has been shown to interact with dipalmitoyl phosphatidylcholine (1) and with histone H1 during the entry of dismantled capsid through the nuclear pore structure (40), whereas no specific cytoplasmic membrane protein receptor has been identified so far, which is compatible with the absence of transmembrane proinflammatory signals that we report here. As exposure of cells to soluble Ad capsid domains allows study of the effect of binding to plasma membrane receptors but does not reproduce the same multistep entry process induced by the whole Ad vector, a question arises as to whether internalization is a necessary step to evoke proinflammatory signals. In this regard, the experiments performed here with the whole Ad.CFTR vector indicate that the block of the interaction between fiber knob and CAR with sCAR-D1, a condition allowing the entry of Ad5 by cooperation of HS-GAGs with {alpha}v integrins and leading to neo-synthesis of viral protein (9), does not induce activation of MAPKs and NF-{kappa}B, thus confirming that penton base and hexon are not involved in these pathways.

HS-GAGs are utilized as attachment receptors for cell entry by different viruses (for review, see reference 3). HS-GAGs are independent of and alternative to CAR for Ad2/5 initial binding (9). The Ad5-derived replication-defective Ad.CFTR vector or Ad full-length fiber, which have the 91KKTK94 sequence for binding of the fiber shaft to HS-GAGs, did not elicit MAPK or NF-{kappa}B activation in A549 cells when binding to CAR was hindered by sCAR-D1, demonstrating the absence of involvement of HS-GAGs in this signaling pathway. A question arises as to whether HS-GAGs play a unique role as "silent" membrane carriers in Ad2/5 entry. Although heparan sulfate proteoglycan-mediated transmembrane signaling has been described during development (for review, see reference 14), our results are compatible with a proposed role as a catalyst of molecular interactions on the surface of the plasma membrane, for instance, as coreceptors for growth factors or concentrating ligands in specific areas of the cell membrane (41).

In conclusion, the observation that the interactions of Ad surface domains elicit early proinflammatory signaling after binding to CAR provides important information concerning future design strategies for new gene transfer vectors. Firstly, modifications of the HI-loop of the fiber knob, which have been developed to retarget Ad vectors to receptors other than CAR (17, 22), may help to reduce the early inflammatory response following interaction of the capsid of helper-dependent gutless Ad vectors with target cells (28). Secondly, gene transfer vectors other than those derived from Ad2/5 that are known to utilize HS-GAGs as receptors may in principle produce a safer application profile for these types of vectors. Examples include adeno-associated parvoviruses type 2, herpes simplex types 1 and 2, and vesicular stomatitis virus G protein-pseudotyped lentiviral vectors (3, 13). Thirdly, while Ad5 full-length fibers alone enter HeLa and 293 cells by binding to HS-GAGs through the 91KKTK94 sequence in the shaft and distribute inside the nucleus by a nuclear localization signal sequence in the tail, binding of fiber to CAR is not followed by internalization (31). Therefore, it may be possible to avoid binding to CAR by modifying the HI-loop of the Ad5 full-length fiber to allow chimeric fibers to be assembled with nucleic acid carriers, thereby establishing effective and safe tools to transfer genes of interest inside the nuclei of specific target cells.


   ACKNOWLEDGMENTS
 
This work was supported by the Italian Cystic Fibrosis Research Foundation (to A.T.), the "Finanziamento Ricerca Fibrosi Cistica 2004, Legge 548/93" (to A.T.), and the Fondazione Cariverona ("Bando 2005—Malattie rare e della povertà") (to G.C).

We are grateful to Roberta Fontana, Marco A. Cassatella, and Federica Calzetti (University of Verona), Antonio Borellini (Microbiological Laboratory, Glaxo Smith Kline), and Maria Cristina Dechecchi (Laboratory of Molecular Pathology) for helpful discussions and suggestions, to Peter Durie (University of Toronto) for critically revising the manuscript, and to Federica Quiri and Valentino Stanzial for excellent technical assistance.


   FOOTNOTES
 
* Corresponding author. Mailing address: Laboratorio Patologia Molecolare, Centro Fibrosi Cistica, Azienda Ospedaliera-Universitaria, Piazzale Stefani 1, 37126 Verona, Italy. Phone: 39 045 807 2364. Fax: 39 045 807 2840. E-mail: giulio.cabrini{at}azosp.vr.it. Back

{triangledown} Published ahead of print on 6 September 2006. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Balakireva, L., G. Schoenhn, E. Thouvenin, and J. Chroboczek. 2003. Binding of adenovirus capsid to dipalmitoyl phosphatidylcholine provides a novel pathway for virus entry. J. Virol. 77:4858-4866.[Abstract/Free Full Text]
  2. Bergelson, J. M., J. A. Cunningham, G. Droguett, E. A. Kurt-Jones, A. Krithivas, J. S. Hong, M. S. Horwitz, R. L. Crowell, and R. W. Finberg. 1997. Isolation of a common receptor for coxsackie B viruses and adenoviruses 2 and 5. Science 275:1320-1323.[Abstract/Free Full Text]
  3. Bomsel, M., and A. Alfsen. 2003. Entry of viruses through the epithelial barrier: pathogenic trickery. Nat. Rev. Mol. Cell Biol. 4:57-68.[CrossRef][Medline]
  4. Borgland, S., G. Bowen, N. Wong, T. Libermann, and D. Muruve. 2000. Adenovirus vector-induced expression of the C-X-C chemokine IP-10 is mediated through capsid-dependent activation of NF-{kappa}B. J. Virol. 74:3941-3947.[Abstract/Free Full Text]
  5. Boulanger, P. A., and F. Puvion. 1973. Large-scale preparation of soluble adenovirus hexon, penton and fiber antigen in highly purified form. Eur. J. Biochem. 39:37-42.[Medline]
  6. Bowen, G., S. Borgland, M. Lam, T. Libermann, N. Wong, and D. A. Muruve. 2002. Adenovirus vector-induced inflammation: capsid-dependent induction of the C_C chemokine RANTES requires NF-{kappa}B. Hum. Gene Ther. 13:367-379.[CrossRef][Medline]
  7. Bruder, J. T., and I. Kovesdi. 1997. Adenovirus infection stimulates the Raf/MAPK signaling pathway and induces interleukin-8 expression. J. Virol. 71:398-404.[Abstract]
  8. Compton, T., E. A. Kurt-Jones, K. W. Boehme, J. Belko, E. Latz, D. T. Golenbock, and R. W. Finberg. 2003. Human cytomegalovirus activates inflammatory cytokine responses via CD14 and Toll-like receptor 2. J. Virol. 77:4588-4596.[Abstract/Free Full Text]
  9. Dechecchi, M. C., P. Melotti, A. Bonizzato, M. Santacatterina, M. Chilosi, and G. Cabrini. 2001. Heparan sulfate glycosaminoglycans are receptors sufficient to mediate the initial binding of adenovirus types 2 and 5. J. Virol. 75:8772-8780.[Abstract/Free Full Text]
  10. Dechecchi, M. C., A. Tamanini, A. Bonizzato, and G. Cabrini. 2000. Heparan sulfate glycosaminoglycans are involved in adenovirus type 5 and 2-host cell interactions. Virology 268:382-390.[CrossRef][Medline]
  11. Freimuth, P., K. Springer, C. Berard, J. Hainfeld, M. Bewley, and J. Flanagan. 1999. Coxsackievirus and adenovirus receptor amino-terminal immunoglobulin V-related domain binds adenovirus type 2 and fiber knob from adenovirus type 12. J. Virol. 73:1392-1398.[Abstract/Free Full Text]
  12. Greber, U. F., M. Willetts, P. Webster, and A. Helenius. 1993. Stepwise dismantling of adenovirus 2 during entry into cells. Cell 75:477-486.[CrossRef][Medline]
  13. Guibinga, G. H., A. Miyanohara, J. D. Esko, and T. Friedmann. 2002. Cell surface heparan sulfate is a receptor for attachment of envelope protein-free retrovirus-like particles and VSV-G pseudotyped MLV-derived retrovirus vectors to target cells. Mol. Ther. 5:538-546.[CrossRef][Medline]
  14. Hacker, U., K. Nybakken, and N. Perrimon. 2005. Heparan sulphate proteoglycans: the sweet side of development. Nat. Rev. Mol. Cell Biol. 6:530-541.[CrossRef][Medline]
  15. Henry, L. J., D. Xia, M. E. Wilke, J. Deisenhofer, and R. D. Gerard. 1994. Characterization of the knob domain of the adenovirus type 5 fiber protein expressed in Escherichia coli. J. Virol. 68:5239-5246.[Abstract/Free Full Text]
  16. Kochanek, S., P. R. Clemens, K. Mitani, H. H. Chen, S. Chan, and C. T. Caskey. 1996. A new adenoviral vector: replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and ß-galactosidase. Proc. Natl. Acad. Sci. USA 93:5731-5736.[Abstract/Free Full Text]
  17. Krasnykh, V., I. Dmitriev, J.-G. Navarro, N. Belousova, E. Kashentseva, J. Xiang, J. T. Douglas, and D. T. Curiel. 2000. Advanced generation adenoviral vectors possess augmented gene transfer efficiency based upon coxsackie adenovirus receptor-independent cellular entry capacity. Cancer Res. 60:6784-6787.[Abstract/Free Full Text]
  18. Li, E., D. Stupack, R. Klemke, D. A. Cheresh, and G. Nemerow. 1998. Adenovirus endocytosis via {alpha}v integrins requires phosphoinositide-3-OH kinase. J. Virol. 72:2055-2061.[Abstract/Free Full Text]
  19. Li, E., D. G. Stupack, S. L. Brown, R. Klemke, D. D. Schaepfer, and G. R. Nemerow. 2000. Association of p130CAS with phosphatidylinositol-3-OH kinase mediates adenovirus cell entry. J. Biol. Chem. 275:14729-14735.[Abstract/Free Full Text]
  20. Liu, Q., and D. A. Muruve. 2003. Molecular basis of the inflammatory response to adenovirus vectors. Gene Ther. 10:935-940.[CrossRef][Medline]
  21. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275.[Free Full Text]
  22. Magnusson, M. K., S. S. Hong, P. Boulanger, and L. Lindholm. 2001. Genetic retargeting of adenovirus:novel strategy employing "deknobbing" of the fiber. J. Virol. 75:7280-7289.[Abstract/Free Full Text]
  23. Melotti, P., E. Nicolis, A. Tamanini, R. Rolfini, A. Pavirani, and G. Cabrini. 2001. Activation of NF-{kappa}B mediates ICAM-1 induction in respiratory cells exposed to an adenovirus-derived vector. Gene Ther. 8:1436-1442.[CrossRef][Medline]
  24. Mittereder, N., K. L. March, and B. C. Trapnell. 1996. Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy. J. Virol. 70:7498-7509.[Abstract]
  25. Morral, N., W. O'Neal, K. Rice, M. Leland, J. Kaplan, P. A. Piedra, H. Zhou, R. J. Parks, R. Velji, E. Aguillar-Cordova, S. Wadsworth, F. L. Graham, S. Kochanek, K. D. Carey, and A. L. Beaudet. 1999. Administration of helper-dependent adenoviral vectors and sequential delivery of different vector serotype for long-term liver-directed gene transfer in baboons. Proc. Natl. Acad. Sci. USA 96:12816-12821.[Abstract/Free Full Text]
  26. Munger, J. S., X. Huang, H. Kawakatsu, M. J. Griffiths, S. L. Dalton, J. Wu, J. F. Pittet, N. Kaminski, C. Garat, M. A. Matthay, D. B. Rifkin, and D. Sheppard. 1999. The intergin alpha v beta 6 binds and activates latent TGF beta 1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell 96:319-328.[CrossRef][Medline]
  27. Muruve, D., M. J. Barnes, I. E. Stillman, and T. Libermann. 1999. Adenoviral gene therapy leads to rapid induction of multiple chemokines and acute neutrophil-dependent hepatic injury in vivo. Hum. Gene Ther. 10:965-976.[CrossRef][Medline]
  28. Muruve, D., M. J. Cotter, A. K. Zaiss, L. R. White, Q. Liu, T. Chan, S. A. Clark, P. J. Ross, R. A. Meulenbroek, G. M. Maelandsmo, and R. J. Parks. 2004. Helper-dependent adenovirus vectors elicit intact innate but attenuated adaptive host immune responses in vivo. J. Virol. 78:5966-5972.[Abstract/Free Full Text]
  29. Peng, Q., D. Lai, T. T. B. Nguyen, V. Chan, T. Matsuda, and S. J. Hirst. 2005. Multiple ß1 integrins mediate enhancement of human airway smooth muscle cytokine secretion by fibronectin and type I collagen. J. Immunol. 174:2258-2264.[Abstract/Free Full Text]
  30. Precious, B., and W. C. Russell. 1995. Growth, purification and titration of adenovirus, p. 193-205. In B. W. J. Mahy (ed.), Virology: a practical approach. IRL Press, Oxford, England.
  31. Rentsendorj, A., H. Agadjanian, X. Chen, M. Cirivello, M. Macveigh, L. Kedes, S. Hamm-Alvarez, and L. K. Medina-Kauwe. 2005. The Ad5 fiber mediates nonviral gene transfer in the absence of the whole virus, utilizing a novel cell entry pathway. Gene Ther. 12:225-237.[CrossRef][Medline]
  32. Roelvink, P. W., G. M. Lee, D. A. Einfeld, I. Kovesdi, and T. J. Wickham. 1999. Identification of a conserved receptor-binding site on the fiber proteins of CAR-recognizing adenoviridae. Science 286:1568-1571.[Abstract/Free Full Text]
  33. Rosenfeld, M. A., K. Yoshimura, B. C. Trapnell, K. Yoneyama, E. R. Rosenthal, W. Dalemans, M. Fukayama, J. Bargon, L. E. Stier, L. Stratford-Perricaudet, M. Perricaudet, W. B. Guggino, A. Pavirani, J.-P. Lecocq, and R. G. Crystal. 1992. In vivo transfer of the human cystic fibrosis transmembrane conductance regulator gene to the airway epithelium. Cell 68:143-155.[CrossRef][Medline]
  34. Sexton, D. W., M. G. Blaylock, and G. M. Walsh. 2001. Human alveolar epithelial cells engulf apoptotic eosinophils by means of integrin- and phosphatidylserine receptor-dependent mechanisms: a process upregulated by dexamethasone. J. Allergy Clin. Immunol. 108:962-969.[CrossRef][Medline]
  35. Smith, T. A. G., N. Idamakanti, M. L. Rollence, J. Marshall-Neff, J. Kim, K. Mulgrew, G. R. Nemerow, M. Kaleko, and S. C. Stevenson. 2003. Adenovirus serotype 5 fiber shaft influences in vivo gene transfer in mice. Hum. Gene Ther. 14:777-787.[CrossRef][Medline]
  36. Soumalainen, M., M. Y. Nakano, K. Boucke, S. Keller, and U. F. Greber. 2001. Adenovirus-activated PKA and p38/MAPK pathways boost microtubule-mediated nuclear targeting of virus. EMBO J. 20:1310-1319.[CrossRef][Medline]
  37. Stilwell, J. L., and R. J. Samulski. 2003. Role of viral vectors and virion shells in cellular gene expression. Mol. Ther. 9:337-346.
  38. Tamanini, A., R. Rolfini, E. Nicolis, P. Melotti, and G. Cabrini. 2003. MAP kinases and NF-{kappa}B collaborate to induce ICAM-1 gene expression in the early phase of adenovirus infection. Virology 307:228-242.[CrossRef][Medline]
  39. Tibbles, L. A., J. C. L. Spurrell, G. Bowen, Q. Liu, M. Lam, A. K. Zaiss, S. M. Robbins, M. D. Hollenberg, T. Wickham, and D. A. Muruve. 2002. Activation of p38 and ERK signaling during adenovirus vector cell entry lead to expression of the C-X-C chemokine IP-10. J. Virol. 76:1559-1568.[Abstract/Free Full Text]
  40. Trotman, L. C., N. Mosberger, M. Fornerod, R. P. Stidwill, and U. F. Greber. 2001. Import of adenovirus DNA involves the nuclear pore complex receptor CAN/Nup214 and histone H1. Nat. Cell Biol. 3:1092-1100.[CrossRef][Medline]
  41. Tumbull, J., A. Powell, and S. Guimond. 2001. Heparan sulfate: decoding a dynamic multifunctional cell regulator. Trends Cell Biol. 11:75-82.[CrossRef][Medline]
  42. Volpers, C., and S. Kochanek. 2004. Adenovirus vectors for gene transfer and therapy. J. Gene Med. 6:S164-S171.[Medline]
  43. Walters, R. W., P. Freimuth, T. O. Moninger, J. Zabner, and M. J. Welsh. 2002. Adenovirus fiber disrupts CAR-mediated intercellular adhesion allowing virus escape. Cell 110:789-799.[CrossRef][Medline]
  44. Wickham, T. J., P. Mathias, D. A. Cheresh, and G. Nemerow. 1993. Integrins {alpha}vß3 and {alpha}vß5 promote adenovirus internalization but not virus attachment. Cell 73:309-319.[CrossRef][Medline]
  45. Yang, Y., Q. Li, H. C. J. Ertl, and J. M. Wilson. 1995. Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J. Virol. 69:2004-2015.[Abstract]

Journal of Virology, November 2006, p. 11241-11254, Vol. 80, No. 22
0022-538X/06/$08.00+0     doi:10.1128/JVI.00721-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
JVI.00721-06v1
80/22/11241    most recent
Right arrow Alert me when this article is cited